human/dist/human.esm.js

/*
  Human
  homepage: <https://github.com/vladmandic/human>
  author: <https://github.com/vladmandic>'
*/

var __defProp = Object.defineProperty;
var __defNormalProp = (obj, key, value) => key in obj ? __defProp(obj, key, { enumerable: true, configurable: true, writable: true, value }) : obj[key] = value;
var __export = (target, all2) => {
  for (var name in all2)
    __defProp(target, name, { get: all2[name], enumerable: true });
};
var __publicField = (obj, key, value) => {
  __defNormalProp(obj, typeof key !== "symbol" ? key + "" : key, value);
  return value;
};
var __accessCheck = (obj, member, msg) => {
  if (!member.has(obj))
    throw TypeError("Cannot " + msg);
};
var __privateGet = (obj, member, getter) => {
  __accessCheck(obj, member, "read from private field");
  return getter ? getter.call(obj) : member.get(obj);
};
var __privateAdd = (obj, member, value) => {
  if (member.has(obj))
    throw TypeError("Cannot add the same private member more than once");
  member instanceof WeakSet ? member.add(obj) : member.set(obj, value);
};
var __privateSet = (obj, member, value, setter) => {
  __accessCheck(obj, member, "write to private field");
  setter ? setter.call(obj, value) : member.set(obj, value);
  return value;
};

// src/util/util.ts
function log(...msg) {
  const dt2 = new Date();
  const ts2 = `${dt2.getHours().toString().padStart(2, "0")}:${dt2.getMinutes().toString().padStart(2, "0")}:${dt2.getSeconds().toString().padStart(2, "0")}.${dt2.getMilliseconds().toString().padStart(3, "0")}`;
  if (msg)
    console.log(ts2, "Human:", ...msg);
}
function join(folder, file) {
  const separator = folder.endsWith("/") ? "" : "/";
  const skipJoin = file.startsWith(".") || file.startsWith("/") || file.startsWith("http:") || file.startsWith("https:") || file.startsWith("file:");
  const path = skipJoin ? `${file}` : `${folder}${separator}${file}`;
  if (!path.toLocaleLowerCase().includes(".json"))
    throw new Error(`modelpath error: expecting json file: ${path}`);
  return path;
}
var now = () => {
  if (typeof performance !== "undefined")
    return performance.now();
  return parseInt((Number(process.hrtime.bigint()) / 1e3 / 1e3).toString());
};
function validate(defaults, config3, parent = "config", msgs = []) {
  for (const key of Object.keys(config3)) {
    if (typeof config3[key] === "object") {
      validate(defaults[key], config3[key], key, msgs);
    } else {
      const defined = defaults && typeof defaults[key] !== "undefined";
      if (!defined)
        msgs.push({ reason: "unknown property", where: `${parent}.${key} = ${config3[key]}` });
      const same = defaults && typeof defaults[key] === typeof config3[key];
      if (defined && !same)
        msgs.push({ reason: "property type mismatch", where: `${parent}.${key} = ${config3[key]}`, expected: typeof defaults[key] });
    }
  }
  if (config3.debug && parent === "config" && msgs.length > 0)
    log("invalid configuration", msgs);
  return msgs;
}
function mergeDeep(...objects) {
  const isObject = (obj) => obj && typeof obj === "object";
  return objects.reduce((prev, obj) => {
    Object.keys(obj || {}).forEach((key) => {
      const pVal = prev[key];
      const oVal = obj[key];
      if (Array.isArray(pVal) && Array.isArray(oVal))
        prev[key] = pVal.concat(...oVal);
      else if (isObject(pVal) && isObject(oVal))
        prev[key] = mergeDeep(pVal, oVal);
      else
        prev[key] = oVal;
    });
    return prev;
  }, {});
}

// src/config.ts
var config = {
  backend: "",
  modelBasePath: "",
  cacheModels: true,
  wasmPath: "",
  wasmPlatformFetch: false,
  debug: true,
  async: true,
  warmup: "full",
  cacheSensitivity: 0.7,
  skipAllowed: false,
  deallocate: false,
  filter: {
    enabled: true,
    equalization: false,
    width: 0,
    height: 0,
    flip: false,
    return: true,
    brightness: 0,
    contrast: 0,
    sharpness: 0,
    blur: 0,
    saturation: 0,
    hue: 0,
    negative: false,
    sepia: false,
    vintage: false,
    kodachrome: false,
    technicolor: false,
    polaroid: false,
    pixelate: 0
  },
  gesture: {
    enabled: true
  },
  face: {
    enabled: true,
    detector: {
      modelPath: "blazeface.json",
      rotation: true,
      maxDetected: 1,
      skipFrames: 99,
      skipTime: 2500,
      minConfidence: 0.2,
      iouThreshold: 0.1,
      mask: false,
      return: false
    },
    mesh: {
      enabled: true,
      modelPath: "facemesh.json"
    },
    attention: {
      enabled: false,
      modelPath: "facemesh-attention.json"
    },
    iris: {
      enabled: true,
      modelPath: "iris.json"
    },
    emotion: {
      enabled: true,
      minConfidence: 0.1,
      skipFrames: 99,
      skipTime: 1500,
      modelPath: "emotion.json"
    },
    description: {
      enabled: true,
      modelPath: "faceres.json",
      skipFrames: 99,
      skipTime: 3e3,
      minConfidence: 0.1
    },
    antispoof: {
      enabled: false,
      skipFrames: 99,
      skipTime: 4e3,
      modelPath: "antispoof.json"
    },
    liveness: {
      enabled: false,
      skipFrames: 99,
      skipTime: 4e3,
      modelPath: "liveness.json"
    }
  },
  body: {
    enabled: true,
    modelPath: "movenet-lightning.json",
    maxDetected: -1,
    minConfidence: 0.3,
    skipFrames: 1,
    skipTime: 200
  },
  hand: {
    enabled: true,
    rotation: true,
    skipFrames: 99,
    skipTime: 1e3,
    minConfidence: 0.5,
    iouThreshold: 0.2,
    maxDetected: -1,
    landmarks: true,
    detector: {
      modelPath: "handtrack.json"
    },
    skeleton: {
      modelPath: "handlandmark-full.json"
    }
  },
  object: {
    enabled: false,
    modelPath: "mb3-centernet.json",
    minConfidence: 0.2,
    iouThreshold: 0.4,
    maxDetected: 10,
    skipFrames: 99,
    skipTime: 2e3
  },
  segmentation: {
    enabled: false,
    modelPath: "selfie.json",
    blur: 8
  }
};

// dist/tfjs.esm.js
var tfjs_esm_exports = {};
__export(tfjs_esm_exports, {
  Abs: () => po,
  Acos: () => ol,
  Acosh: () => ul,
  AdadeltaOptimizer: () => kb,
  AdagradOptimizer: () => Sb,
  AdamOptimizer: () => Ib,
  AdamaxOptimizer: () => Cb,
  Add: () => Sr,
  AddN: () => Sa,
  All: () => ll,
  Any: () => cl,
  ArgMax: () => Ia,
  ArgMin: () => dl,
  Asin: () => pl,
  Asinh: () => hl,
  Atan: () => fl,
  Atan2: () => gl,
  Atanh: () => ml,
  AvgPool: () => Ca,
  AvgPool3D: () => Jd,
  AvgPool3DGrad: () => hg,
  AvgPoolGrad: () => pg,
  BackendWasm: () => Zde,
  BatchMatMul: () => Na,
  BatchToSpaceND: () => ho,
  Bincount: () => fg,
  BroadcastArgs: () => mg,
  BroadcastTo: () => M$,
  Callback: () => WW,
  CallbackList: () => rB,
  Cast: () => Ta,
  Ceil: () => $a,
  ClipByValue: () => Ir,
  Complex: () => ep,
  ComplexAbs: () => tp,
  Concat: () => fo,
  Conv2D: () => _a,
  Conv2DBackpropFilter: () => gg,
  Conv2DBackpropInput: () => Aa,
  Conv3D: () => np,
  Conv3DBackpropFilterV2: () => bg,
  Conv3DBackpropInputV2: () => yg,
  Cos: () => Ea,
  Cosh: () => Ra,
  CropAndResize: () => go,
  Cumprod: () => mo,
  Cumsum: () => Da,
  CustomCallback: () => oB,
  DataStorage: () => Yd,
  DenseBincount: () => vg,
  DepthToSpace: () => bo,
  DepthwiseConv2dNative: () => Fa,
  DepthwiseConv2dNativeBackpropFilter: () => xg,
  DepthwiseConv2dNativeBackpropInput: () => wg,
  Diag: () => kg,
  Dilation2D: () => sp,
  Dilation2DBackpropFilter: () => sm,
  Dilation2DBackpropInput: () => nm,
  ENV: () => lk,
  EarlyStopping: () => UW,
  Einsum: () => rp,
  Elu: () => Pa,
  EluGrad: () => Sg,
  Environment: () => R$,
  Equal: () => yo,
  Erf: () => bl,
  Exp: () => za,
  ExpandDims: () => vo,
  Expm1: () => xo,
  FFT: () => Ig,
  Fill: () => yl,
  FlipLeftRight: () => wo,
  Floor: () => Ma,
  FloorDiv: () => La,
  FromPixels: () => yd,
  FusedBatchNorm: () => Ba,
  FusedConv2D: () => ia,
  FusedDepthwiseConv2D: () => oa,
  GPGPUContext: () => em,
  GatherNd: () => So,
  GatherV2: () => ko,
  GraphModel: () => K4,
  Greater: () => Io,
  GreaterEqual: () => Va,
  History: () => iB,
  IFFT: () => Cg,
  Identity: () => Wa,
  Imag: () => ap,
  InputSpec: () => Ft,
  IsFinite: () => vl,
  IsInf: () => xl,
  IsNan: () => wl,
  KernelBackend: () => il,
  LRN: () => op,
  LRNGrad: () => Tg,
  LayerVariable: () => vz,
  LayersModel: () => dr,
  LeakyRelu: () => Ua,
  Less: () => Co,
  LessEqual: () => No,
  LinSpace: () => Ng,
  Log: () => Ga,
  Log1p: () => kl,
  LogSoftmax: () => L$,
  LogicalAnd: () => To,
  LogicalNot: () => Sl,
  LogicalOr: () => ip,
  LowerBound: () => rpe,
  MathBackendCPU: () => Y0,
  MathBackendWebGL: () => K1,
  Max: () => Ha,
  MaxPool: () => ja,
  MaxPool3D: () => up,
  MaxPool3DGrad: () => _g,
  MaxPoolGrad: () => $g,
  MaxPoolWithArgmax: () => Ag,
  Maximum: () => qa,
  Mean: () => Ka,
  Min: () => Xa,
  Minimum: () => Ya,
  MirrorPad: () => Qa,
  Mod: () => Il,
  MomentumOptimizer: () => Nb,
  Multinomial: () => Eg,
  Multiply: () => Za,
  Neg: () => $o,
  NonMaxSuppressionV3: () => Ao,
  NonMaxSuppressionV4: () => Cl,
  NonMaxSuppressionV5: () => Eo,
  NotEqual: () => _o,
  OP_SCOPE_SUFFIX: () => b_,
  OneHot: () => Do,
  OnesLike: () => Ro,
  Optimizer: () => _r,
  OptimizerConstructors: () => Ur,
  Pack: () => Fo,
  PadV2: () => Ja,
  Pool: () => ape,
  Pow: () => ei,
  Prelu: () => ti,
  Prod: () => ni,
  RMSPropOptimizer: () => Tb,
  RNN: () => Ar,
  Range: () => Nl,
  Rank: () => u_,
  Real: () => lp,
  RealDiv: () => Oa,
  Reciprocal: () => Tl,
  Reduction: () => vO,
  Relu: () => si,
  Relu6: () => ai,
  Reshape: () => Oo,
  ResizeBilinear: () => ri,
  ResizeBilinearGrad: () => Dg,
  ResizeNearestNeighbor: () => $l,
  ResizeNearestNeighborGrad: () => Rg,
  Reverse: () => Po,
  RotateWithOffset: () => Yo,
  Round: () => zo,
  Rsqrt: () => ii,
  SGDOptimizer: () => Ap,
  ScatterNd: () => Mo,
  SearchSorted: () => Fg,
  Select: () => Lo,
  Selu: () => _l,
  Sequential: () => Qb,
  Sigmoid: () => ui,
  Sign: () => Al,
  Sin: () => oi,
  Sinh: () => Vo,
  Slice: () => Bo,
  Softmax: () => di,
  Softplus: () => El,
  SpaceToBatchND: () => Wo,
  SparseFillEmptyRows: () => cp,
  SparseReshape: () => Rl,
  SparseSegmentMean: () => dp,
  SparseSegmentSum: () => pp,
  SparseToDense: () => hp,
  SplitV: () => Uo,
  Sqrt: () => li,
  Square: () => Dl,
  SquaredDifference: () => pi,
  Step: () => gi,
  StridedSlice: () => Go,
  StringNGrams: () => fp,
  StringSplit: () => Og,
  StringToHashBucketFast: () => Pg,
  Sub: () => hi,
  Sum: () => ci,
  SymbolicTensor: () => $s,
  Tan: () => Ho,
  Tanh: () => fi,
  Tensor: () => et,
  TensorBuffer: () => Wt,
  Tile: () => Cr,
  TopK: () => qo,
  Transform: () => jo,
  Transpose: () => mi,
  Unique: () => zg,
  Unpack: () => Ko,
  UnsortedSegmentSum: () => mp,
  UpperBound: () => ipe,
  Variable: () => xd,
  ZerosLike: () => Xo,
  _FusedMatMul: () => aa,
  abs: () => Lt,
  acos: () => ZA,
  acosh: () => eE,
  add: () => ie,
  addN: () => nE,
  all: () => nS,
  any: () => ym,
  argMax: () => Xu,
  argMin: () => oE,
  asin: () => lE,
  asinh: () => dE,
  atan: () => hE,
  atan2: () => mE,
  atanh: () => bE,
  avgPool: () => Qg,
  avgPool3d: () => iS,
  backend: () => HA,
  backend_util: () => C,
  basicLSTMCell: () => Ipe,
  batchNorm: () => Qu,
  batchNorm2d: () => PE,
  batchNorm3d: () => ME,
  batchNorm4d: () => BE,
  batchToSpaceND: () => Zg,
  bincount: () => oS,
  booleanMaskAsync: () => Zpe,
  broadcastArgs: () => UE,
  broadcastTo: () => ad,
  broadcast_util: () => Qo,
  browser: () => Pk,
  buffer: () => Ae,
  callbacks: () => uhe,
  cast: () => le,
  ceil: () => qE,
  clipByValue: () => Vn,
  clone: () => ur,
  complex: () => ua,
  concat: () => Ot,
  concat1d: () => XE,
  concat2d: () => QE,
  concat3d: () => JE,
  concat4d: () => tR,
  constraints: () => OL,
  conv1d: () => uS,
  conv2d: () => da,
  conv2dTranspose: () => lS,
  conv3d: () => cS,
  conv3dTranspose: () => lR,
  copyRegisteredKernels: () => lpe,
  cos: () => eb,
  cosh: () => pS,
  cosineWindow: () => zS,
  cumprod: () => xm,
  cumsum: () => hS,
  customGrad: () => qs,
  data: () => X4,
  denseBincount: () => mR,
  deprecationWarn: () => eS,
  depthToSpace: () => bR,
  depthwiseConv2d: () => xp,
  deregisterOp: () => che,
  device_util: () => yp,
  diag: () => Cpe,
  dilation2d: () => wR,
  disableDeprecationWarnings: () => hpe,
  dispose: () => De,
  disposeVariables: () => fpe,
  div: () => xe,
  divNoNan: () => NR,
  dot: () => Npe,
  dropout: () => fF,
  einsum: () => _R,
  elu: () => wp,
  enableDebugMode: () => ppe,
  enableProdMode: () => dpe,
  enclosingPowerOfTwo: () => mF,
  engine: () => ds,
  env: () => K,
  equal: () => Xn,
  erf: () => RR,
  euclideanNorm: () => UR,
  exp: () => Yn,
  expandDims: () => Pn,
  expm1: () => jR,
  eye: () => yS,
  fft: () => bb,
  fill: () => Ll,
  findBackend: () => wpe,
  findBackendFactory: () => kpe,
  floor: () => kp,
  floorDiv: () => tS,
  forceHalfFloat: () => h8,
  fused: () => fa,
  gather: () => Zu,
  gatherND: () => dF,
  gather_util: () => Mk,
  getBackend: () => vpe,
  getGradient: () => lx,
  getKernel: () => rm,
  getKernelsForBackend: () => am,
  getThreadsCount: () => xhe,
  gpgpu_util: () => eX,
  grad: () => _pe,
  grads: () => Ape,
  greater: () => Un,
  greaterEqual: () => Zo,
  ifft: () => Td,
  imag: () => rb,
  image: () => jn,
  inTopKAsync: () => ehe,
  initializers: () => BL,
  input: () => ZB,
  io: () => An,
  irfft: () => RS,
  isFinite: () => Tpe,
  isInf: () => $pe,
  isNaN: () => rD,
  keep: () => qt,
  kernel_impls: () => ws,
  layers: () => nB,
  leakyRelu: () => ab,
  less: () => vS,
  lessEqual: () => Jo,
  linalg: () => JO,
  linspace: () => uD,
  loadGraphModel: () => dhe,
  loadLayersModel: () => ihe,
  localResponseNormalization: () => cD,
  log: () => Qn,
  log1p: () => ib,
  logSigmoid: () => Dpe,
  logSoftmax: () => xS,
  logSumExp: () => xD,
  logicalAnd: () => Ds,
  logicalNot: () => ob,
  logicalOr: () => wS,
  logicalXor: () => Fpe,
  losses: () => she,
  lowerBound: () => ND,
  matMul: () => Ve,
  math: () => fA,
  max: () => As,
  maxPool: () => ub,
  maxPool3d: () => SS,
  maxPoolWithArgmax: () => AD,
  maximum: () => $r,
  mean: () => It,
  memory: () => bm,
  meshgrid: () => Ope,
  metrics: () => wW,
  min: () => wm,
  minimum: () => Ip,
  mirrorPad: () => OD,
  mod: () => zD,
  model: () => rhe,
  models: () => zW,
  moments: () => lb,
  movingAverage: () => Jpe,
  mul: () => V,
  multiRNNCell: () => Ppe,
  multinomial: () => VD,
  neg: () => kt,
  nextFrame: () => HS,
  norm: () => sb,
  notEqual: () => Ju,
  oneHot: () => Sd,
  ones: () => Mn,
  onesLike: () => Zn,
  op: () => L,
  outerProduct: () => zpe,
  pad: () => bi,
  pad1d: () => Mpe,
  pad2d: () => Lpe,
  pad3d: () => Bpe,
  pad4d: () => Vpe,
  pool: () => Wpe,
  pow: () => ha,
  prelu: () => db,
  print: () => Q_,
  prod: () => IS,
  profile: () => mpe,
  rand: () => Upe,
  randomGamma: () => Gpe,
  randomNormal: () => o3,
  randomUniform: () => Vl,
  range: () => el,
  ready: () => ype,
  real: () => Cd,
  reciprocal: () => d3,
  registerBackend: () => vp,
  registerCallbackConstructor: () => ohe,
  registerGradient: () => V$,
  registerKernel: () => Fl,
  registerOp: () => lhe,
  regularizers: () => MW,
  relu: () => Xs,
  relu6: () => CS,
  removeBackend: () => xpe,
  reshape: () => U,
  reverse: () => Jn,
  reverse1d: () => Hpe,
  reverse2d: () => qpe,
  reverse3d: () => jpe,
  reverse4d: () => Kpe,
  rfft: () => yb,
  round: () => NS,
  rsqrt: () => TS,
  scalar: () => we,
  scatterND: () => oF,
  scatter_util: () => Bk,
  searchSorted: () => kS,
  selu: () => $S,
  separableConv2d: () => S3,
  sequential: () => ahe,
  serialization: () => re,
  setBackend: () => bpe,
  setPlatform: () => Spe,
  setThreadsCount: () => vhe,
  setWasmPath: () => bhe,
  setWasmPaths: () => yhe,
  setWebGLContext: () => q5,
  setdiff1dAsync: () => C3,
  shared: () => iv,
  sigmoid: () => Hs,
  sign: () => T3,
  signal: () => nhe,
  sin: () => _S,
  sinh: () => AS,
  slice: () => qe,
  slice1d: () => fb,
  slice2d: () => ES,
  slice3d: () => mb,
  slice4d: () => Nd,
  slice_util: () => wt,
  softmax: () => gb,
  softplus: () => Bl,
  spaceToBatchND: () => cb,
  sparse: () => Hc,
  sparseToDense: () => PS,
  spectral: () => the,
  split: () => Bn,
  sqrt: () => dn,
  square: () => ct,
  squaredDifference: () => DS,
  squeeze: () => mr,
  stack: () => es,
  step: () => Cp,
  stridedSlice: () => H3,
  string: () => Hf,
  sub: () => ge,
  sum: () => ve,
  sumOutType: () => bp,
  tan: () => j3,
  tanh: () => Yu,
  tensor: () => ms,
  tensor1d: () => Zt,
  tensor2d: () => Zi,
  tensor3d: () => xA,
  tensor4d: () => Xpe,
  tensor5d: () => Ype,
  tensor6d: () => Qpe,
  tensor_util: () => _s,
  test_util: () => zA,
  tidy: () => j,
  tile: () => hs,
  time: () => gpe,
  topk: () => X3,
  train: () => Li,
  transpose: () => Ge,
  truncatedNormal: () => vb,
  unique: () => xx,
  unregisterGradient: () => upe,
  unregisterKernel: () => ope,
  unsortedSegmentSum: () => J3,
  unstack: () => Fs,
  upcastType: () => cn,
  upperBound: () => tF,
  util: () => w,
  valueAndGrad: () => Epe,
  valueAndGrads: () => Rpe,
  variable: () => nF,
  variableGrads: () => hD,
  version: () => khe,
  version_converter: () => phe,
  version_core: () => cpe,
  version_cpu: () => hhe,
  version_layers: () => vI,
  version_wasm: () => whe,
  version_webgl: () => fhe,
  webgl: () => mhe,
  webgl_util: () => H5,
  webgpu: () => Foe,
  where: () => vn,
  whereAsync: () => OS,
  zeros: () => $t,
  zerosLike: () => je
});
var jT = Object.create;
var Kd = Object.defineProperty;
var KT = Object.getOwnPropertyDescriptor;
var Qw = Object.getOwnPropertyNames;
var XT = Object.getPrototypeOf;
var YT = Object.prototype.hasOwnProperty;
var QT = (e) => Kd(e, "__esModule", { value: true });
var Mt = (e, t) => function() {
  return t || (0, e[Qw(e)[0]])((t = { exports: {} }).exports, t), t.exports;
};
var Ee = (e, t) => {
  for (var n in t)
    Kd(e, n, { get: t[n], enumerable: true });
};
var ZT = (e, t, n, s) => {
  if (t && typeof t == "object" || typeof t == "function")
    for (let r of Qw(t))
      !YT.call(e, r) && (n || r !== "default") && Kd(e, r, { get: () => t[r], enumerable: !(s = KT(t, r)) || s.enumerable });
  return e;
};
var wa = (e, t) => ZT(QT(Kd(e != null ? jT(XT(e)) : {}, "default", !t && e && e.__esModule ? { get: () => e.default, enumerable: true } : { value: e, enumerable: true })), e);
var JT = Mt({ "src/node_modules/long/src/long.js"(e, t) {
  t.exports = s;
  var n = null;
  try {
    n = new WebAssembly.Instance(new WebAssembly.Module(new Uint8Array([0, 97, 115, 109, 1, 0, 0, 0, 1, 13, 2, 96, 0, 1, 127, 96, 4, 127, 127, 127, 127, 1, 127, 3, 7, 6, 0, 1, 1, 1, 1, 1, 6, 6, 1, 127, 1, 65, 0, 11, 7, 50, 6, 3, 109, 117, 108, 0, 1, 5, 100, 105, 118, 95, 115, 0, 2, 5, 100, 105, 118, 95, 117, 0, 3, 5, 114, 101, 109, 95, 115, 0, 4, 5, 114, 101, 109, 95, 117, 0, 5, 8, 103, 101, 116, 95, 104, 105, 103, 104, 0, 0, 10, 191, 1, 6, 4, 0, 35, 0, 11, 36, 1, 1, 126, 32, 0, 173, 32, 1, 173, 66, 32, 134, 132, 32, 2, 173, 32, 3, 173, 66, 32, 134, 132, 126, 34, 4, 66, 32, 135, 167, 36, 0, 32, 4, 167, 11, 36, 1, 1, 126, 32, 0, 173, 32, 1, 173, 66, 32, 134, 132, 32, 2, 173, 32, 3, 173, 66, 32, 134, 132, 127, 34, 4, 66, 32, 135, 167, 36, 0, 32, 4, 167, 11, 36, 1, 1, 126, 32, 0, 173, 32, 1, 173, 66, 32, 134, 132, 32, 2, 173, 32, 3, 173, 66, 32, 134, 132, 128, 34, 4, 66, 32, 135, 167, 36, 0, 32, 4, 167, 11, 36, 1, 1, 126, 32, 0, 173, 32, 1, 173, 66, 32, 134, 132, 32, 2, 173, 32, 3, 173, 66, 32, 134, 132, 129, 34, 4, 66, 32, 135, 167, 36, 0, 32, 4, 167, 11, 36, 1, 1, 126, 32, 0, 173, 32, 1, 173, 66, 32, 134, 132, 32, 2, 173, 32, 3, 173, 66, 32, 134, 132, 130, 34, 4, 66, 32, 135, 167, 36, 0, 32, 4, 167, 11])), {}).exports;
  } catch (O) {
  }
  function s(O, T, z) {
    this.low = O | 0, this.high = T | 0, this.unsigned = !!z;
  }
  s.prototype.__isLong__, Object.defineProperty(s.prototype, "__isLong__", { value: true });
  function r(O) {
    return (O && O.__isLong__) === true;
  }
  s.isLong = r;
  var a = {}, i = {};
  function o(O, T) {
    var z, W, q;
    return T ? (O >>>= 0, (q = 0 <= O && O < 256) && (W = i[O], W) ? W : (z = l(O, (O | 0) < 0 ? -1 : 0, true), q && (i[O] = z), z)) : (O |= 0, (q = -128 <= O && O < 128) && (W = a[O], W) ? W : (z = l(O, O < 0 ? -1 : 0, false), q && (a[O] = z), z));
  }
  s.fromInt = o;
  function u(O, T) {
    if (isNaN(O))
      return T ? x : v;
    if (T) {
      if (O < 0)
        return x;
      if (O >= g)
        return E;
    } else {
      if (O <= -b)
        return P;
      if (O + 1 >= b)
        return R;
    }
    return O < 0 ? u(-O, T).neg() : l(O % m | 0, O / m | 0, T);
  }
  s.fromNumber = u;
  function l(O, T, z) {
    return new s(O, T, z);
  }
  s.fromBits = l;
  var c = Math.pow;
  function p(O, T, z) {
    if (O.length === 0)
      throw Error("empty string");
    if (O === "NaN" || O === "Infinity" || O === "+Infinity" || O === "-Infinity")
      return v;
    if (typeof T == "number" ? (z = T, T = false) : T = !!T, z = z || 10, z < 2 || 36 < z)
      throw RangeError("radix");
    var W;
    if ((W = O.indexOf("-")) > 0)
      throw Error("interior hyphen");
    if (W === 0)
      return p(O.substring(1), T, z).neg();
    for (var q = u(c(z, 8)), X = v, Y = 0; Y < O.length; Y += 8) {
      var Z = Math.min(8, O.length - Y), te = parseInt(O.substring(Y, Y + Z), z);
      if (Z < 8) {
        var J = u(c(z, Z));
        X = X.mul(J).add(u(te));
      } else
        X = X.mul(q), X = X.add(u(te));
    }
    return X.unsigned = T, X;
  }
  s.fromString = p;
  function d(O, T) {
    return typeof O == "number" ? u(O, T) : typeof O == "string" ? p(O, T) : l(O.low, O.high, typeof T == "boolean" ? T : O.unsigned);
  }
  s.fromValue = d;
  var h = 1 << 16, f = 1 << 24, m = h * h, g = m * m, b = g / 2, y = o(f), v = o(0);
  s.ZERO = v;
  var x = o(0, true);
  s.UZERO = x;
  var k = o(1);
  s.ONE = k;
  var I = o(1, true);
  s.UONE = I;
  var $ = o(-1);
  s.NEG_ONE = $;
  var R = l(-1, 2147483647, false);
  s.MAX_VALUE = R;
  var E = l(-1, -1, true);
  s.MAX_UNSIGNED_VALUE = E;
  var P = l(0, -2147483648, false);
  s.MIN_VALUE = P;
  var A = s.prototype;
  A.toInt = function() {
    return this.unsigned ? this.low >>> 0 : this.low;
  }, A.toNumber = function() {
    return this.unsigned ? (this.high >>> 0) * m + (this.low >>> 0) : this.high * m + (this.low >>> 0);
  }, A.toString = function(T) {
    if (T = T || 10, T < 2 || 36 < T)
      throw RangeError("radix");
    if (this.isZero())
      return "0";
    if (this.isNegative())
      if (this.eq(P)) {
        var z = u(T), W = this.div(z), q = W.mul(z).sub(this);
        return W.toString(T) + q.toInt().toString(T);
      } else
        return "-" + this.neg().toString(T);
    for (var X = u(c(T, 6), this.unsigned), Y = this, Z = ""; ; ) {
      var te = Y.div(X), J = Y.sub(te.mul(X)).toInt() >>> 0, se = J.toString(T);
      if (Y = te, Y.isZero())
        return se + Z;
      for (; se.length < 6; )
        se = "0" + se;
      Z = "" + se + Z;
    }
  }, A.getHighBits = function() {
    return this.high;
  }, A.getHighBitsUnsigned = function() {
    return this.high >>> 0;
  }, A.getLowBits = function() {
    return this.low;
  }, A.getLowBitsUnsigned = function() {
    return this.low >>> 0;
  }, A.getNumBitsAbs = function() {
    if (this.isNegative())
      return this.eq(P) ? 64 : this.neg().getNumBitsAbs();
    for (var T = this.high != 0 ? this.high : this.low, z = 31; z > 0 && (T & 1 << z) == 0; z--)
      ;
    return this.high != 0 ? z + 33 : z + 1;
  }, A.isZero = function() {
    return this.high === 0 && this.low === 0;
  }, A.eqz = A.isZero, A.isNegative = function() {
    return !this.unsigned && this.high < 0;
  }, A.isPositive = function() {
    return this.unsigned || this.high >= 0;
  }, A.isOdd = function() {
    return (this.low & 1) === 1;
  }, A.isEven = function() {
    return (this.low & 1) === 0;
  }, A.equals = function(T) {
    return r(T) || (T = d(T)), this.unsigned !== T.unsigned && this.high >>> 31 === 1 && T.high >>> 31 === 1 ? false : this.high === T.high && this.low === T.low;
  }, A.eq = A.equals, A.notEquals = function(T) {
    return !this.eq(T);
  }, A.neq = A.notEquals, A.ne = A.notEquals, A.lessThan = function(T) {
    return this.comp(T) < 0;
  }, A.lt = A.lessThan, A.lessThanOrEqual = function(T) {
    return this.comp(T) <= 0;
  }, A.lte = A.lessThanOrEqual, A.le = A.lessThanOrEqual, A.greaterThan = function(T) {
    return this.comp(T) > 0;
  }, A.gt = A.greaterThan, A.greaterThanOrEqual = function(T) {
    return this.comp(T) >= 0;
  }, A.gte = A.greaterThanOrEqual, A.ge = A.greaterThanOrEqual, A.compare = function(T) {
    if (r(T) || (T = d(T)), this.eq(T))
      return 0;
    var z = this.isNegative(), W = T.isNegative();
    return z && !W ? -1 : !z && W ? 1 : this.unsigned ? T.high >>> 0 > this.high >>> 0 || T.high === this.high && T.low >>> 0 > this.low >>> 0 ? -1 : 1 : this.sub(T).isNegative() ? -1 : 1;
  }, A.comp = A.compare, A.negate = function() {
    return !this.unsigned && this.eq(P) ? P : this.not().add(k);
  }, A.neg = A.negate, A.add = function(T) {
    r(T) || (T = d(T));
    var z = this.high >>> 16, W = this.high & 65535, q = this.low >>> 16, X = this.low & 65535, Y = T.high >>> 16, Z = T.high & 65535, te = T.low >>> 16, J = T.low & 65535, se = 0, ne = 0, oe = 0, ae = 0;
    return ae += X + J, oe += ae >>> 16, ae &= 65535, oe += q + te, ne += oe >>> 16, oe &= 65535, ne += W + Z, se += ne >>> 16, ne &= 65535, se += z + Y, se &= 65535, l(oe << 16 | ae, se << 16 | ne, this.unsigned);
  }, A.subtract = function(T) {
    return r(T) || (T = d(T)), this.add(T.neg());
  }, A.sub = A.subtract, A.multiply = function(T) {
    if (this.isZero())
      return v;
    if (r(T) || (T = d(T)), n) {
      var z = n.mul(this.low, this.high, T.low, T.high);
      return l(z, n.get_high(), this.unsigned);
    }
    if (T.isZero())
      return v;
    if (this.eq(P))
      return T.isOdd() ? P : v;
    if (T.eq(P))
      return this.isOdd() ? P : v;
    if (this.isNegative())
      return T.isNegative() ? this.neg().mul(T.neg()) : this.neg().mul(T).neg();
    if (T.isNegative())
      return this.mul(T.neg()).neg();
    if (this.lt(y) && T.lt(y))
      return u(this.toNumber() * T.toNumber(), this.unsigned);
    var W = this.high >>> 16, q = this.high & 65535, X = this.low >>> 16, Y = this.low & 65535, Z = T.high >>> 16, te = T.high & 65535, J = T.low >>> 16, se = T.low & 65535, ne = 0, oe = 0, ae = 0, de = 0;
    return de += Y * se, ae += de >>> 16, de &= 65535, ae += X * se, oe += ae >>> 16, ae &= 65535, ae += Y * J, oe += ae >>> 16, ae &= 65535, oe += q * se, ne += oe >>> 16, oe &= 65535, oe += X * J, ne += oe >>> 16, oe &= 65535, oe += Y * te, ne += oe >>> 16, oe &= 65535, ne += W * se + q * J + X * te + Y * Z, ne &= 65535, l(ae << 16 | de, ne << 16 | oe, this.unsigned);
  }, A.mul = A.multiply, A.divide = function(T) {
    if (r(T) || (T = d(T)), T.isZero())
      throw Error("division by zero");
    if (n) {
      if (!this.unsigned && this.high === -2147483648 && T.low === -1 && T.high === -1)
        return this;
      var z = (this.unsigned ? n.div_u : n.div_s)(this.low, this.high, T.low, T.high);
      return l(z, n.get_high(), this.unsigned);
    }
    if (this.isZero())
      return this.unsigned ? x : v;
    var W, q, X;
    if (this.unsigned) {
      if (T.unsigned || (T = T.toUnsigned()), T.gt(this))
        return x;
      if (T.gt(this.shru(1)))
        return I;
      X = x;
    } else {
      if (this.eq(P)) {
        if (T.eq(k) || T.eq($))
          return P;
        if (T.eq(P))
          return k;
        var Y = this.shr(1);
        return W = Y.div(T).shl(1), W.eq(v) ? T.isNegative() ? k : $ : (q = this.sub(T.mul(W)), X = W.add(q.div(T)), X);
      } else if (T.eq(P))
        return this.unsigned ? x : v;
      if (this.isNegative())
        return T.isNegative() ? this.neg().div(T.neg()) : this.neg().div(T).neg();
      if (T.isNegative())
        return this.div(T.neg()).neg();
      X = v;
    }
    for (q = this; q.gte(T); ) {
      W = Math.max(1, Math.floor(q.toNumber() / T.toNumber()));
      for (var Z = Math.ceil(Math.log(W) / Math.LN2), te = Z <= 48 ? 1 : c(2, Z - 48), J = u(W), se = J.mul(T); se.isNegative() || se.gt(q); )
        W -= te, J = u(W, this.unsigned), se = J.mul(T);
      J.isZero() && (J = k), X = X.add(J), q = q.sub(se);
    }
    return X;
  }, A.div = A.divide, A.modulo = function(T) {
    if (r(T) || (T = d(T)), n) {
      var z = (this.unsigned ? n.rem_u : n.rem_s)(this.low, this.high, T.low, T.high);
      return l(z, n.get_high(), this.unsigned);
    }
    return this.sub(this.div(T).mul(T));
  }, A.mod = A.modulo, A.rem = A.modulo, A.not = function() {
    return l(~this.low, ~this.high, this.unsigned);
  }, A.and = function(T) {
    return r(T) || (T = d(T)), l(this.low & T.low, this.high & T.high, this.unsigned);
  }, A.or = function(T) {
    return r(T) || (T = d(T)), l(this.low | T.low, this.high | T.high, this.unsigned);
  }, A.xor = function(T) {
    return r(T) || (T = d(T)), l(this.low ^ T.low, this.high ^ T.high, this.unsigned);
  }, A.shiftLeft = function(T) {
    return r(T) && (T = T.toInt()), (T &= 63) === 0 ? this : T < 32 ? l(this.low << T, this.high << T | this.low >>> 32 - T, this.unsigned) : l(0, this.low << T - 32, this.unsigned);
  }, A.shl = A.shiftLeft, A.shiftRight = function(T) {
    return r(T) && (T = T.toInt()), (T &= 63) === 0 ? this : T < 32 ? l(this.low >>> T | this.high << 32 - T, this.high >> T, this.unsigned) : l(this.high >> T - 32, this.high >= 0 ? 0 : -1, this.unsigned);
  }, A.shr = A.shiftRight, A.shiftRightUnsigned = function(T) {
    if (r(T) && (T = T.toInt()), T &= 63, T === 0)
      return this;
    var z = this.high;
    if (T < 32) {
      var W = this.low;
      return l(W >>> T | z << 32 - T, z >>> T, this.unsigned);
    } else
      return T === 32 ? l(z, 0, this.unsigned) : l(z >>> T - 32, 0, this.unsigned);
  }, A.shru = A.shiftRightUnsigned, A.shr_u = A.shiftRightUnsigned, A.toSigned = function() {
    return this.unsigned ? l(this.low, this.high, false) : this;
  }, A.toUnsigned = function() {
    return this.unsigned ? this : l(this.low, this.high, true);
  }, A.toBytes = function(T) {
    return T ? this.toBytesLE() : this.toBytesBE();
  }, A.toBytesLE = function() {
    var T = this.high, z = this.low;
    return [z & 255, z >>> 8 & 255, z >>> 16 & 255, z >>> 24, T & 255, T >>> 8 & 255, T >>> 16 & 255, T >>> 24];
  }, A.toBytesBE = function() {
    var T = this.high, z = this.low;
    return [T >>> 24, T >>> 16 & 255, T >>> 8 & 255, T & 255, z >>> 24, z >>> 16 & 255, z >>> 8 & 255, z & 255];
  }, s.fromBytes = function(T, z, W) {
    return W ? s.fromBytesLE(T, z) : s.fromBytesBE(T, z);
  }, s.fromBytesLE = function(T, z) {
    return new s(T[0] | T[1] << 8 | T[2] << 16 | T[3] << 24, T[4] | T[5] << 8 | T[6] << 16 | T[7] << 24, z);
  }, s.fromBytesBE = function(T, z) {
    return new s(T[4] << 24 | T[5] << 16 | T[6] << 8 | T[7], T[0] << 24 | T[1] << 16 | T[2] << 8 | T[3], z);
  };
} });
var e$ = Mt({ "(disabled):src/node_modules/node-fetch/browser.js"() {
} });
var t$ = Mt({ "(disabled):util"() {
} });
var n$ = Mt({ "src/node_modules/seedrandom/lib/alea.js"(e, t) {
  (function(n, s, r) {
    function a(l) {
      var c = this, p = u();
      c.next = function() {
        var d = 2091639 * c.s0 + c.c * 23283064365386963e-26;
        return c.s0 = c.s1, c.s1 = c.s2, c.s2 = d - (c.c = d | 0);
      }, c.c = 1, c.s0 = p(" "), c.s1 = p(" "), c.s2 = p(" "), c.s0 -= p(l), c.s0 < 0 && (c.s0 += 1), c.s1 -= p(l), c.s1 < 0 && (c.s1 += 1), c.s2 -= p(l), c.s2 < 0 && (c.s2 += 1), p = null;
    }
    function i(l, c) {
      return c.c = l.c, c.s0 = l.s0, c.s1 = l.s1, c.s2 = l.s2, c;
    }
    function o(l, c) {
      var p = new a(l), d = c && c.state, h = p.next;
      return h.int32 = function() {
        return p.next() * 4294967296 | 0;
      }, h.double = function() {
        return h() + (h() * 2097152 | 0) * 11102230246251565e-32;
      }, h.quick = h, d && (typeof d == "object" && i(d, p), h.state = function() {
        return i(p, {});
      }), h;
    }
    function u() {
      var l = 4022871197, c = function(p) {
        p = String(p);
        for (var d = 0; d < p.length; d++) {
          l += p.charCodeAt(d);
          var h = 0.02519603282416938 * l;
          l = h >>> 0, h -= l, h *= l, l = h >>> 0, h -= l, l += h * 4294967296;
        }
        return (l >>> 0) * 23283064365386963e-26;
      };
      return c;
    }
    s && s.exports ? s.exports = o : r && r.amd ? r(function() {
      return o;
    }) : this.alea = o;
  })(e, typeof t == "object" && t, typeof define == "function" && define);
} });
var s$ = Mt({ "src/node_modules/seedrandom/lib/xor128.js"(e, t) {
  (function(n, s, r) {
    function a(u) {
      var l = this, c = "";
      l.x = 0, l.y = 0, l.z = 0, l.w = 0, l.next = function() {
        var d = l.x ^ l.x << 11;
        return l.x = l.y, l.y = l.z, l.z = l.w, l.w ^= l.w >>> 19 ^ d ^ d >>> 8;
      }, u === (u | 0) ? l.x = u : c += u;
      for (var p = 0; p < c.length + 64; p++)
        l.x ^= c.charCodeAt(p) | 0, l.next();
    }
    function i(u, l) {
      return l.x = u.x, l.y = u.y, l.z = u.z, l.w = u.w, l;
    }
    function o(u, l) {
      var c = new a(u), p = l && l.state, d = function() {
        return (c.next() >>> 0) / 4294967296;
      };
      return d.double = function() {
        do
          var h = c.next() >>> 11, f = (c.next() >>> 0) / 4294967296, m = (h + f) / (1 << 21);
        while (m === 0);
        return m;
      }, d.int32 = c.next, d.quick = d, p && (typeof p == "object" && i(p, c), d.state = function() {
        return i(c, {});
      }), d;
    }
    s && s.exports ? s.exports = o : r && r.amd ? r(function() {
      return o;
    }) : this.xor128 = o;
  })(e, typeof t == "object" && t, typeof define == "function" && define);
} });
var r$ = Mt({ "src/node_modules/seedrandom/lib/xorwow.js"(e, t) {
  (function(n, s, r) {
    function a(u) {
      var l = this, c = "";
      l.next = function() {
        var d = l.x ^ l.x >>> 2;
        return l.x = l.y, l.y = l.z, l.z = l.w, l.w = l.v, (l.d = l.d + 362437 | 0) + (l.v = l.v ^ l.v << 4 ^ (d ^ d << 1)) | 0;
      }, l.x = 0, l.y = 0, l.z = 0, l.w = 0, l.v = 0, u === (u | 0) ? l.x = u : c += u;
      for (var p = 0; p < c.length + 64; p++)
        l.x ^= c.charCodeAt(p) | 0, p == c.length && (l.d = l.x << 10 ^ l.x >>> 4), l.next();
    }
    function i(u, l) {
      return l.x = u.x, l.y = u.y, l.z = u.z, l.w = u.w, l.v = u.v, l.d = u.d, l;
    }
    function o(u, l) {
      var c = new a(u), p = l && l.state, d = function() {
        return (c.next() >>> 0) / 4294967296;
      };
      return d.double = function() {
        do
          var h = c.next() >>> 11, f = (c.next() >>> 0) / 4294967296, m = (h + f) / (1 << 21);
        while (m === 0);
        return m;
      }, d.int32 = c.next, d.quick = d, p && (typeof p == "object" && i(p, c), d.state = function() {
        return i(c, {});
      }), d;
    }
    s && s.exports ? s.exports = o : r && r.amd ? r(function() {
      return o;
    }) : this.xorwow = o;
  })(e, typeof t == "object" && t, typeof define == "function" && define);
} });
var a$ = Mt({ "src/node_modules/seedrandom/lib/xorshift7.js"(e, t) {
  (function(n, s, r) {
    function a(u) {
      var l = this;
      l.next = function() {
        var p = l.x, d = l.i, h, f, m;
        return h = p[d], h ^= h >>> 7, f = h ^ h << 24, h = p[d + 1 & 7], f ^= h ^ h >>> 10, h = p[d + 3 & 7], f ^= h ^ h >>> 3, h = p[d + 4 & 7], f ^= h ^ h << 7, h = p[d + 7 & 7], h = h ^ h << 13, f ^= h ^ h << 9, p[d] = f, l.i = d + 1 & 7, f;
      };
      function c(p, d) {
        var h, f, m = [];
        if (d === (d | 0))
          f = m[0] = d;
        else
          for (d = "" + d, h = 0; h < d.length; ++h)
            m[h & 7] = m[h & 7] << 15 ^ d.charCodeAt(h) + m[h + 1 & 7] << 13;
        for (; m.length < 8; )
          m.push(0);
        for (h = 0; h < 8 && m[h] === 0; ++h)
          ;
        for (h == 8 ? f = m[7] = -1 : f = m[h], p.x = m, p.i = 0, h = 256; h > 0; --h)
          p.next();
      }
      c(l, u);
    }
    function i(u, l) {
      return l.x = u.x.slice(), l.i = u.i, l;
    }
    function o(u, l) {
      u == null && (u = +new Date());
      var c = new a(u), p = l && l.state, d = function() {
        return (c.next() >>> 0) / 4294967296;
      };
      return d.double = function() {
        do
          var h = c.next() >>> 11, f = (c.next() >>> 0) / 4294967296, m = (h + f) / (1 << 21);
        while (m === 0);
        return m;
      }, d.int32 = c.next, d.quick = d, p && (p.x && i(p, c), d.state = function() {
        return i(c, {});
      }), d;
    }
    s && s.exports ? s.exports = o : r && r.amd ? r(function() {
      return o;
    }) : this.xorshift7 = o;
  })(e, typeof t == "object" && t, typeof define == "function" && define);
} });
var i$ = Mt({ "src/node_modules/seedrandom/lib/xor4096.js"(e, t) {
  (function(n, s, r) {
    function a(u) {
      var l = this;
      l.next = function() {
        var p = l.w, d = l.X, h = l.i, f, m;
        return l.w = p = p + 1640531527 | 0, m = d[h + 34 & 127], f = d[h = h + 1 & 127], m ^= m << 13, f ^= f << 17, m ^= m >>> 15, f ^= f >>> 12, m = d[h] = m ^ f, l.i = h, m + (p ^ p >>> 16) | 0;
      };
      function c(p, d) {
        var h, f, m, g, b, y = [], v = 128;
        for (d === (d | 0) ? (f = d, d = null) : (d = d + "\0", f = 0, v = Math.max(v, d.length)), m = 0, g = -32; g < v; ++g)
          d && (f ^= d.charCodeAt((g + 32) % d.length)), g === 0 && (b = f), f ^= f << 10, f ^= f >>> 15, f ^= f << 4, f ^= f >>> 13, g >= 0 && (b = b + 1640531527 | 0, h = y[g & 127] ^= f + b, m = h == 0 ? m + 1 : 0);
        for (m >= 128 && (y[(d && d.length || 0) & 127] = -1), m = 127, g = 4 * 128; g > 0; --g)
          f = y[m + 34 & 127], h = y[m = m + 1 & 127], f ^= f << 13, h ^= h << 17, f ^= f >>> 15, h ^= h >>> 12, y[m] = f ^ h;
        p.w = b, p.X = y, p.i = m;
      }
      c(l, u);
    }
    function i(u, l) {
      return l.i = u.i, l.w = u.w, l.X = u.X.slice(), l;
    }
    function o(u, l) {
      u == null && (u = +new Date());
      var c = new a(u), p = l && l.state, d = function() {
        return (c.next() >>> 0) / 4294967296;
      };
      return d.double = function() {
        do
          var h = c.next() >>> 11, f = (c.next() >>> 0) / 4294967296, m = (h + f) / (1 << 21);
        while (m === 0);
        return m;
      }, d.int32 = c.next, d.quick = d, p && (p.X && i(p, c), d.state = function() {
        return i(c, {});
      }), d;
    }
    s && s.exports ? s.exports = o : r && r.amd ? r(function() {
      return o;
    }) : this.xor4096 = o;
  })(e, typeof t == "object" && t, typeof define == "function" && define);
} });
var o$ = Mt({ "src/node_modules/seedrandom/lib/tychei.js"(e, t) {
  (function(n, s, r) {
    function a(u) {
      var l = this, c = "";
      l.next = function() {
        var d = l.b, h = l.c, f = l.d, m = l.a;
        return d = d << 25 ^ d >>> 7 ^ h, h = h - f | 0, f = f << 24 ^ f >>> 8 ^ m, m = m - d | 0, l.b = d = d << 20 ^ d >>> 12 ^ h, l.c = h = h - f | 0, l.d = f << 16 ^ h >>> 16 ^ m, l.a = m - d | 0;
      }, l.a = 0, l.b = 0, l.c = -1640531527, l.d = 1367130551, u === Math.floor(u) ? (l.a = u / 4294967296 | 0, l.b = u | 0) : c += u;
      for (var p = 0; p < c.length + 20; p++)
        l.b ^= c.charCodeAt(p) | 0, l.next();
    }
    function i(u, l) {
      return l.a = u.a, l.b = u.b, l.c = u.c, l.d = u.d, l;
    }
    function o(u, l) {
      var c = new a(u), p = l && l.state, d = function() {
        return (c.next() >>> 0) / 4294967296;
      };
      return d.double = function() {
        do
          var h = c.next() >>> 11, f = (c.next() >>> 0) / 4294967296, m = (h + f) / (1 << 21);
        while (m === 0);
        return m;
      }, d.int32 = c.next, d.quick = d, p && (typeof p == "object" && i(p, c), d.state = function() {
        return i(c, {});
      }), d;
    }
    s && s.exports ? s.exports = o : r && r.amd ? r(function() {
      return o;
    }) : this.tychei = o;
  })(e, typeof t == "object" && t, typeof define == "function" && define);
} });
var u$ = Mt({ "(disabled):crypto"() {
} });
var l$ = Mt({ "src/node_modules/seedrandom/seedrandom.js"(e, t) {
  (function(n, s, r) {
    var a = 256, i = 6, o = 52, u = "random", l = r.pow(a, i), c = r.pow(2, o), p = c * 2, d = a - 1, h;
    function f(k, I, $) {
      var R = [];
      I = I == true ? { entropy: true } : I || {};
      var E = y(b(I.entropy ? [k, x(s)] : k == null ? v() : k, 3), R), P = new m(R), A = function() {
        for (var O = P.g(i), T = l, z = 0; O < c; )
          O = (O + z) * a, T *= a, z = P.g(1);
        for (; O >= p; )
          O /= 2, T /= 2, z >>>= 1;
        return (O + z) / T;
      };
      return A.int32 = function() {
        return P.g(4) | 0;
      }, A.quick = function() {
        return P.g(4) / 4294967296;
      }, A.double = A, y(x(P.S), s), (I.pass || $ || function(O, T, z, W) {
        return W && (W.S && g(W, P), O.state = function() {
          return g(P, {});
        }), z ? (r[u] = O, T) : O;
      })(A, E, "global" in I ? I.global : this == r, I.state);
    }
    function m(k) {
      var I, $ = k.length, R = this, E = 0, P = R.i = R.j = 0, A = R.S = [];
      for ($ || (k = [$++]); E < a; )
        A[E] = E++;
      for (E = 0; E < a; E++)
        A[E] = A[P = d & P + k[E % $] + (I = A[E])], A[P] = I;
      (R.g = function(O) {
        for (var T, z = 0, W = R.i, q = R.j, X = R.S; O--; )
          T = X[W = d & W + 1], z = z * a + X[d & (X[W] = X[q = d & q + T]) + (X[q] = T)];
        return R.i = W, R.j = q, z;
      })(a);
    }
    function g(k, I) {
      return I.i = k.i, I.j = k.j, I.S = k.S.slice(), I;
    }
    function b(k, I) {
      var $ = [], R = typeof k, E;
      if (I && R == "object")
        for (E in k)
          try {
            $.push(b(k[E], I - 1));
          } catch (P) {
          }
      return $.length ? $ : R == "string" ? k : k + "\0";
    }
    function y(k, I) {
      for (var $ = k + "", R, E = 0; E < $.length; )
        I[d & E] = d & (R ^= I[d & E] * 19) + $.charCodeAt(E++);
      return x(I);
    }
    function v() {
      try {
        var k;
        return h && (k = h.randomBytes) ? k = k(a) : (k = new Uint8Array(a), (n.crypto || n.msCrypto).getRandomValues(k)), x(k);
      } catch (R) {
        var I = n.navigator, $ = I && I.plugins;
        return [+new Date(), n, $, n.screen, x(s)];
      }
    }
    function x(k) {
      return String.fromCharCode.apply(0, k);
    }
    if (y(r.random(), s), typeof t == "object" && t.exports) {
      t.exports = f;
      try {
        h = u$();
      } catch (k) {
      }
    } else
      typeof define == "function" && define.amd ? define(function() {
        return f;
      }) : r["seed" + u] = f;
  })(typeof self != "undefined" ? self : e, [], Math);
} });
var Xd = Mt({ "src/node_modules/seedrandom/index.js"(e, t) {
  var n = n$(), s = s$(), r = r$(), a = a$(), i = i$(), o = o$(), u = l$();
  u.alea = n, u.xor128 = s, u.xorwow = r, u.xorshift7 = a, u.xor4096 = i, u.tychei = o, t.exports = u;
} });
var Zw = Mt({ "(disabled):src/node_modules/string_decoder/index.js"() {
} });
var og = Mt({ "(disabled):fs"() {
} });
var md = Mt({ "(disabled):path"() {
} });
var c$ = Mt({ "(disabled):worker_threads"() {
} });
var d$ = Mt({ "(disabled):perf_hooks"() {
} });
var p$ = Mt({ "(disabled):os"() {
} });
var h$ = Mt({ "src/tfjs-backend-wasm/wasm-out/tfjs-backend-wasm-threaded-simd.js"(e, t) {
  var n = (() => {
    var s = typeof document != "undefined" && document.currentScript ? document.currentScript.src : void 0;
    return typeof __filename != "undefined" && (s = s || __filename), function(r) {
      r = r || {};
      function a() {
        return Ce.buffer != nn && rs(Ce.buffer), uc;
      }
      function i() {
        return Ce.buffer != nn && rs(Ce.buffer), lc;
      }
      function o() {
        return Ce.buffer != nn && rs(Ce.buffer), bu;
      }
      function u() {
        return Ce.buffer != nn && rs(Ce.buffer), cc;
      }
      function l() {
        return Ce.buffer != nn && rs(Ce.buffer), dc;
      }
      function c() {
        return Ce.buffer != nn && rs(Ce.buffer), pc;
      }
      function p() {
        return Ce.buffer != nn && rs(Ce.buffer), hc;
      }
      var d = typeof r != "undefined" ? r : {}, h, f;
      d.ready = new Promise(function(N, D) {
        h = N, f = D;
      });
      var m;
      typeof process != "undefined" && process.listeners && (m = { uncaughtException: process.listeners("uncaughtException"), unhandledRejection: process.listeners("unhandledRejection") });
      var g = Object.assign({}, d), b = [], y = "./this.program", v = (N, D) => {
        throw D;
      }, x = typeof window == "object", k = typeof importScripts == "function", I = typeof process == "object" && typeof process.versions == "object" && typeof process.versions.node == "string", $ = d.ENVIRONMENT_IS_PTHREAD || false, R = "";
      function E(N) {
        return d.locateFile ? d.locateFile(N, R) : R + N;
      }
      var P, A, O, T;
      function z(N) {
        if (N instanceof Nu)
          return;
        J("exiting due to exception: " + N);
      }
      var W, q, X;
      if (I) {
        k ? R = md().dirname(R) + "/" : R = __dirname + "/", X = () => {
          q || (W = og(), q = md());
        }, P = function(B, Q) {
          return X(), B = q.normalize(B), W.readFileSync(B, Q ? void 0 : "utf8");
        }, O = (D) => {
          var B = P(D, true);
          return B.buffer || (B = new Uint8Array(B)), B;
        }, A = (D, B, Q) => {
          X(), D = q.normalize(D), W.readFile(D, function(ue, pe) {
            ue ? Q(ue) : B(pe.buffer);
          });
        }, process.argv.length > 1 && (y = process.argv[1].replace(/\\/g, "/")), b = process.argv.slice(2), process.on("uncaughtException", function(D) {
          if (!(D instanceof Nu))
            throw D;
        }), process.on("unhandledRejection", function(D) {
          throw D;
        }), v = (D, B) => {
          if (zr())
            throw process.exitCode = D, B;
          z(B), process.exit(D);
        }, d.inspect = function() {
          return "[Emscripten Module object]";
        };
        let N;
        try {
          N = c$();
        } catch (D) {
          throw console.error('The "worker_threads" module is not supported in this node.js build - perhaps a newer version is needed?'), D;
        }
        global.Worker = N.Worker;
      } else
        (x || k) && (k ? R = self.location.href : typeof document != "undefined" && document.currentScript && (R = document.currentScript.src), typeof s != "undefined" && s && (R = s), R.indexOf("blob:") !== 0 ? R = R.substr(0, R.replace(/[?#].*/, "").lastIndexOf("/") + 1) : R = "", I || (P = (N) => {
          var D = new XMLHttpRequest();
          return D.open("GET", N, false), D.send(null), D.responseText;
        }, k && (O = (N) => {
          var D = new XMLHttpRequest();
          return D.open("GET", N, false), D.responseType = "arraybuffer", D.send(null), new Uint8Array(D.response);
        }), A = (N, D, B) => {
          var Q = new XMLHttpRequest();
          Q.open("GET", N, true), Q.responseType = "arraybuffer", Q.onload = () => {
            if (Q.status == 200 || Q.status == 0 && Q.response) {
              D(Q.response);
              return;
            }
            B();
          }, Q.onerror = B, Q.send(null);
        }), T = (N) => document.title = N);
      I && typeof performance == "undefined" && (global.performance = d$().performance);
      var Y = console.log.bind(console), Z = console.warn.bind(console);
      I && (X(), Y = (N) => W.writeSync(1, N + `
`), Z = (N) => W.writeSync(2, N + `
`));
      var te = d.print || Y, J = d.printErr || Z;
      Object.assign(d, g), g = null, d.arguments && (b = d.arguments), d.thisProgram && (y = d.thisProgram), d.quit && (v = d.quit);
      var se = 4;
      function ne(N) {
        ne.shown || (ne.shown = {}), ne.shown[N] || (ne.shown[N] = 1, J(N));
      }
      function oe(N, D) {
        if (typeof WebAssembly.Function == "function") {
          for (var B = { i: "i32", j: "i64", f: "f32", d: "f64" }, Q = { parameters: [], results: D[0] == "v" ? [] : [B[D[0]]] }, ue = 1; ue < D.length; ++ue)
            Q.parameters.push(B[D[ue]]);
          return new WebAssembly.Function(Q, N);
        }
        var pe = [1, 0, 1, 96], ye = D.slice(0, 1), Te = D.slice(1), bt = { i: 127, j: 126, f: 125, d: 124 };
        pe.push(Te.length);
        for (var ue = 0; ue < Te.length; ++ue)
          pe.push(bt[Te[ue]]);
        ye == "v" ? pe.push(0) : pe = pe.concat([1, bt[ye]]), pe[1] = pe.length - 2;
        var us = new Uint8Array([0, 97, 115, 109, 1, 0, 0, 0].concat(pe, [2, 7, 1, 1, 101, 1, 102, 0, 0, 7, 5, 1, 1, 102, 0, 0])), ls = new WebAssembly.Module(us), Wc = new WebAssembly.Instance(ls, { e: { f: N } }), Tu = Wc.exports.f;
        return Tu;
      }
      var ae = [], de;
      function me() {
        if (ae.length)
          return ae.pop();
        try {
          Fn.grow(1);
        } catch (N) {
          throw N instanceof RangeError ? "Unable to grow wasm table. Set ALLOW_TABLE_GROWTH." : N;
        }
        return Fn.length - 1;
      }
      function ke(N, D) {
        for (var B = N; B < N + D; B++) {
          var Q = Ei(B);
          Q && de.set(Q, B);
        }
      }
      var Ie = 0, Re = (N) => {
        Ie = N;
      }, Pe = Atomics.load, Xe = Atomics.store, Je = Atomics.compareExchange, Ye;
      d.wasmBinary && (Ye = d.wasmBinary);
      var tt = d.noExitRuntime || true;
      typeof WebAssembly != "object" && $i("no native wasm support detected");
      var Ce, ut, at = false, Jt;
      function Nt(N, D) {
        N || $i(D);
      }
      function In(N) {
        var D = d["_" + N];
        return D;
      }
      function Rt(N, D, B, Q, ue) {
        var pe = { string: function(Tn) {
          var Mi = 0;
          if (Tn != null && Tn !== 0) {
            var ox = (Tn.length << 2) + 1;
            Mi = zi(ox), Ms(Tn, Mi, ox);
          }
          return Mi;
        }, array: function(Tn) {
          var Mi = zi(Tn.length);
          return Ls(Tn, Mi), Mi;
        } };
        function ye(Tn) {
          return D === "string" ? tn(Tn) : D === "boolean" ? Boolean(Tn) : Tn;
        }
        var Te = In(N), bt = [], us = 0;
        if (Q)
          for (var ls = 0; ls < Q.length; ls++) {
            var Wc = pe[B[ls]];
            Wc ? (us === 0 && (us = Mf()), bt[ls] = Wc(Q[ls])) : bt[ls] = Q[ls];
          }
        var Tu = Te.apply(null, bt);
        function qT(Tn) {
          return us !== 0 && Mc(us), ye(Tn);
        }
        return Tu = qT(Tu), Tu;
      }
      function en(N, D, B, Q) {
        B = B || [];
        var ue = B.every(function(ye) {
          return ye === "number";
        }), pe = D !== "string";
        return pe && ue && !Q ? In(N) : function() {
          return Rt(N, D, B, arguments, Q);
        };
      }
      var Cn = 1;
      function Nn(N) {
        var D = new TextDecoder(N);
        this.decode = (B) => (B.buffer instanceof SharedArrayBuffer && (B = new Uint8Array(B)), D.decode.call(D, B));
      }
      var Yt = typeof TextDecoder != "undefined" ? new Nn("utf8") : void 0;
      function Dn(N, D, B) {
        for (var Q = D + B, ue = D; N[ue] && !(ue >= Q); )
          ++ue;
        if (ue - D > 16 && N.subarray && Yt)
          return Yt.decode(N.subarray(D, ue));
        for (var pe = ""; D < ue; ) {
          var ye = N[D++];
          if (!(ye & 128)) {
            pe += String.fromCharCode(ye);
            continue;
          }
          var Te = N[D++] & 63;
          if ((ye & 224) == 192) {
            pe += String.fromCharCode((ye & 31) << 6 | Te);
            continue;
          }
          var bt = N[D++] & 63;
          if ((ye & 240) == 224 ? ye = (ye & 15) << 12 | Te << 6 | bt : ye = (ye & 7) << 18 | Te << 12 | bt << 6 | N[D++] & 63, ye < 65536)
            pe += String.fromCharCode(ye);
          else {
            var us = ye - 65536;
            pe += String.fromCharCode(55296 | us >> 10, 56320 | us & 1023);
          }
        }
        return pe;
      }
      function tn(N, D) {
        return N ? Dn(i(), N, D) : "";
      }
      function zs(N, D, B, Q) {
        if (!(Q > 0))
          return 0;
        for (var ue = B, pe = B + Q - 1, ye = 0; ye < N.length; ++ye) {
          var Te = N.charCodeAt(ye);
          if (Te >= 55296 && Te <= 57343) {
            var bt = N.charCodeAt(++ye);
            Te = 65536 + ((Te & 1023) << 10) | bt & 1023;
          }
          if (Te <= 127) {
            if (B >= pe)
              break;
            D[B++] = Te;
          } else if (Te <= 2047) {
            if (B + 1 >= pe)
              break;
            D[B++] = 192 | Te >> 6, D[B++] = 128 | Te & 63;
          } else if (Te <= 65535) {
            if (B + 2 >= pe)
              break;
            D[B++] = 224 | Te >> 12, D[B++] = 128 | Te >> 6 & 63, D[B++] = 128 | Te & 63;
          } else {
            if (B + 3 >= pe)
              break;
            D[B++] = 240 | Te >> 18, D[B++] = 128 | Te >> 12 & 63, D[B++] = 128 | Te >> 6 & 63, D[B++] = 128 | Te & 63;
          }
        }
        return D[B] = 0, B - ue;
      }
      function Ms(N, D, B) {
        return zs(N, i(), D, B);
      }
      function Ci(N) {
        for (var D = 0, B = 0; B < N.length; ++B) {
          var Q = N.charCodeAt(B);
          Q >= 55296 && Q <= 57343 && (Q = 65536 + ((Q & 1023) << 10) | N.charCodeAt(++B) & 1023), Q <= 127 ? ++D : Q <= 2047 ? D += 2 : Q <= 65535 ? D += 3 : D += 4;
        }
        return D;
      }
      var Zs = typeof TextDecoder != "undefined" ? new Nn("utf-16le") : void 0;
      function Ls(N, D) {
        a().set(N, D);
      }
      function gu(N, D, B) {
        for (var Q = 0; Q < N.length; ++Q)
          a()[D++ >> 0] = N.charCodeAt(Q);
        B || (a()[D >> 0] = 0);
      }
      function Ni(N, D) {
        return N % D > 0 && (N += D - N % D), N;
      }
      var nn, uc, lc, bu, cc, dc, Vv, pc, hc;
      $ && (nn = d.buffer);
      function rs(N) {
        nn = N, d.HEAP8 = uc = new Int8Array(N), d.HEAP16 = bu = new Int16Array(N), d.HEAP32 = dc = new Int32Array(N), d.HEAPU8 = lc = new Uint8Array(N), d.HEAPU16 = cc = new Uint16Array(N), d.HEAPU32 = Vv = new Uint32Array(N), d.HEAPF32 = pc = new Float32Array(N), d.HEAPF64 = hc = new Float64Array(N);
      }
      var fc = d.INITIAL_MEMORY || 16777216;
      if ($)
        Ce = d.wasmMemory, nn = d.buffer;
      else if (d.wasmMemory)
        Ce = d.wasmMemory;
      else if (Ce = new WebAssembly.Memory({ initial: fc / 65536, maximum: 32768, shared: true }), !(Ce.buffer instanceof SharedArrayBuffer))
        throw J("requested a shared WebAssembly.Memory but the returned buffer is not a SharedArrayBuffer, indicating that while the browser has SharedArrayBuffer it does not have WebAssembly threads support - you may need to set a flag"), I && console.log("(on node you may need: --experimental-wasm-threads --experimental-wasm-bulk-memory and also use a recent version)"), Error("bad memory");
      Ce && (nn = Ce.buffer), fc = nn.byteLength, rs(nn);
      var Fn, Ti = [], Js = [], uh = [], mc = [], Pr = false, lh = false, gc = 0;
      function zr() {
        return tt || gc > 0;
      }
      function sn() {
        if (d.preRun)
          for (typeof d.preRun == "function" && (d.preRun = [d.preRun]); d.preRun.length; )
            Wv(d.preRun.shift());
        wc(Ti);
      }
      function yu() {
        Pr = true, !$ && wc(Js);
      }
      function ch() {
        $ || ($e.terminateAllThreads(), lh = true);
      }
      function dh() {
        if (!$) {
          if (d.postRun)
            for (typeof d.postRun == "function" && (d.postRun = [d.postRun]); d.postRun.length; )
              vu(d.postRun.shift());
          wc(mc);
        }
      }
      function Wv(N) {
        Ti.unshift(N);
      }
      function Uv(N) {
        Js.unshift(N);
      }
      function vu(N) {
        mc.unshift(N);
      }
      var er = 0, bc = null, as = null;
      function xu(N) {
        er++, d.monitorRunDependencies && d.monitorRunDependencies(er);
      }
      function Gv(N) {
        if (er--, d.monitorRunDependencies && d.monitorRunDependencies(er), er == 0 && (bc !== null && (clearInterval(bc), bc = null), as)) {
          var D = as;
          as = null, D();
        }
      }
      d.preloadedImages = {}, d.preloadedAudios = {};
      function $i(N) {
        $ ? postMessage({ cmd: "onAbort", arg: N }) : d.onAbort && d.onAbort(N), N = "Aborted(" + N + ")", J(N), at = true, Jt = 1, N += ". Build with -s ASSERTIONS=1 for more info.";
        var D = new WebAssembly.RuntimeError(N);
        throw f(D), D;
      }
      var ph = "data:application/octet-stream;base64,";
      function yc(N) {
        return N.startsWith(ph);
      }
      function vc(N) {
        return N.startsWith("file://");
      }
      var rn;
      rn = "tfjs-backend-wasm-threaded-simd.wasm", yc(rn) || (rn = E(rn));
      function xc(N) {
        try {
          if (N == rn && Ye)
            return new Uint8Array(Ye);
          if (O)
            return O(N);
          throw "both async and sync fetching of the wasm failed";
        } catch (D) {
          $i(D);
        }
      }
      function _i() {
        if (!Ye && (x || k)) {
          if (typeof fetch == "function" && !vc(rn))
            return fetch(rn, { credentials: "same-origin" }).then(function(N) {
              if (!N.ok)
                throw "failed to load wasm binary file at '" + rn + "'";
              return N.arrayBuffer();
            }).catch(function() {
              return xc(rn);
            });
          if (A)
            return new Promise(function(N, D) {
              A(rn, function(B) {
                N(new Uint8Array(B));
              }, D);
            });
        }
        return Promise.resolve().then(function() {
          return xc(rn);
        });
      }
      function hh() {
        var N = { env: Dc, wasi_snapshot_preview1: Dc };
        function D(ye, Te) {
          var bt = ye.exports;
          if (d.asm = bt, xh(d.asm.emscripten_tls_init), Fn = d.asm.__indirect_function_table, Uv(d.asm.__wasm_call_ctors), ut = Te, !$) {
            var us = $e.unusedWorkers.length;
            $e.unusedWorkers.forEach(function(ls) {
              $e.loadWasmModuleToWorker(ls, function() {
                --us || Gv("wasm-instantiate");
              });
            });
          }
        }
        $ || xu("wasm-instantiate");
        function B(ye) {
          D(ye.instance, ye.module);
        }
        function Q(ye) {
          return _i().then(function(Te) {
            return WebAssembly.instantiate(Te, N);
          }).then(function(Te) {
            return Te;
          }).then(ye, function(Te) {
            J("failed to asynchronously prepare wasm: " + Te), $i(Te);
          });
        }
        function ue() {
          return !Ye && typeof WebAssembly.instantiateStreaming == "function" && !yc(rn) && !vc(rn) && typeof fetch == "function" ? fetch(rn, { credentials: "same-origin" }).then(function(ye) {
            var Te = WebAssembly.instantiateStreaming(ye, N);
            return Te.then(B, function(bt) {
              return J("wasm streaming compile failed: " + bt), J("falling back to ArrayBuffer instantiation"), Q(B);
            });
          }) : Q(B);
        }
        if (d.instantiateWasm)
          try {
            var pe = d.instantiateWasm(N, D);
            return pe;
          } catch (ye) {
            return J("Module.instantiateWasm callback failed with error: " + ye), false;
          }
        return ue().catch(f), {};
      }
      var Hv, qv, fh = {};
      function wc(N) {
        for (; N.length > 0; ) {
          var D = N.shift();
          if (typeof D == "function") {
            D(d);
            continue;
          }
          var B = D.func;
          typeof B == "number" ? D.arg === void 0 ? Ei(B)() : Ei(B)(D.arg) : B(D.arg === void 0 ? null : D.arg);
        }
      }
      function Ai(N) {
        var D = Mf(), B = N();
        return Mc(D), B;
      }
      function eT(N) {
        return N;
      }
      function jv(N) {
        var D = /\b_Z[\w\d_]+/g;
        return N.replace(D, function(B) {
          var Q = B;
          return B === Q ? B : Q + " [" + B + "]";
        });
      }
      function mh(N) {
        l()[N >> 2] = 0;
        var D = $e.pthreads[N];
        delete $e.pthreads[N], D.worker.terminate(), zf(N), $e.runningWorkers.splice($e.runningWorkers.indexOf(D.worker), 1), D.worker.pthread = void 0;
      }
      function gh(N) {
        var D = $e.pthreads[N];
        D.worker.postMessage({ cmd: "cancel" });
      }
      function kc(N) {
        var D = $e.pthreads[N];
        if (D) {
          l()[N >> 2] = 0;
          var B = D.worker;
          $e.returnWorkerToPool(B);
        }
      }
      function Sc(N) {
        UT(N);
      }
      function bh(N) {
        if (N instanceof Nu || N == "unwind")
          return Jt;
        v(1, N);
      }
      var $e = { unusedWorkers: [], runningWorkers: [], tlsInitFunctions: [], init: function() {
        $ ? $e.initWorker() : $e.initMainThread();
      }, initMainThread: function() {
        for (var N = 8, D = 0; D < N; ++D)
          $e.allocateUnusedWorker();
      }, initWorker: function() {
        tt = false;
      }, pthreads: {}, setExitStatus: function(N) {
        Jt = N;
      }, terminateAllThreads: function() {
        for (var N in $e.pthreads) {
          var D = $e.pthreads[N];
          D && D.worker && $e.returnWorkerToPool(D.worker);
        }
        for (var B = 0; B < $e.unusedWorkers.length; ++B) {
          var Q = $e.unusedWorkers[B];
          Q.terminate();
        }
        $e.unusedWorkers = [];
      }, returnWorkerToPool: function(N) {
        $e.runWithoutMainThreadQueuedCalls(function() {
          delete $e.pthreads[N.pthread.threadInfoStruct], $e.unusedWorkers.push(N), $e.runningWorkers.splice($e.runningWorkers.indexOf(N), 1), zf(N.pthread.threadInfoStruct), N.pthread = void 0;
        });
      }, runWithoutMainThreadQueuedCalls: function(N) {
        l()[ix >> 2] = 0;
        try {
          N();
        } finally {
          l()[ix >> 2] = 1;
        }
      }, receiveObjectTransfer: function(N) {
      }, threadInit: function() {
        for (var N in $e.tlsInitFunctions)
          $e.tlsInitFunctions[N]();
      }, loadWasmModuleToWorker: function(N, D) {
        N.onmessage = (B) => {
          var Q = B.data, ue = Q.cmd;
          if (N.pthread && ($e.currentProxiedOperationCallerThread = N.pthread.threadInfoStruct), Q.targetThread && Q.targetThread != zc()) {
            var pe = $e.pthreads[Q.targetThread];
            pe ? pe.worker.postMessage(Q, Q.transferList) : J('Internal error! Worker sent a message "' + ue + '" to target pthread ' + Q.targetThread + ", but that thread no longer exists!"), $e.currentProxiedOperationCallerThread = void 0;
            return;
          }
          ue === "processQueuedMainThreadWork" ? tx() : ue === "spawnThread" ? Cc(Q) : ue === "cleanupThread" ? kc(Q.thread) : ue === "killThread" ? mh(Q.thread) : ue === "cancelThread" ? gh(Q.thread) : ue === "loaded" ? (N.loaded = true, D && D(N), N.runPthread && (N.runPthread(), delete N.runPthread)) : ue === "print" ? te("Thread " + Q.threadId + ": " + Q.text) : ue === "printErr" ? J("Thread " + Q.threadId + ": " + Q.text) : ue === "alert" ? alert("Thread " + Q.threadId + ": " + Q.text) : Q.target === "setimmediate" ? N.postMessage(Q) : ue === "onAbort" ? d.onAbort && d.onAbort(Q.arg) : J("worker sent an unknown command " + ue), $e.currentProxiedOperationCallerThread = void 0;
        }, N.onerror = (B) => {
          var Q = "worker sent an error!";
          throw J(Q + " " + B.filename + ":" + B.lineno + ": " + B.message), B;
        }, I && (N.on("message", function(B) {
          N.onmessage({ data: B });
        }), N.on("error", function(B) {
          N.onerror(B);
        }), N.on("detachedExit", function() {
        })), N.postMessage({ cmd: "load", urlOrBlob: d.mainScriptUrlOrBlob || s, wasmMemory: Ce, wasmModule: ut });
      }, allocateUnusedWorker: function() {
        var N = E("tfjs-backend-wasm-threaded-simd.worker.js");
        $e.unusedWorkers.push(new Worker(N));
      }, getNewWorker: function() {
        return $e.unusedWorkers.length == 0 && ($e.allocateUnusedWorker(), $e.loadWasmModuleToWorker($e.unusedWorkers[0])), $e.unusedWorkers.pop();
      } };
      function yh() {
        var N = zc(), D = l()[N + 44 >> 2], B = l()[N + 48 >> 2], Q = D - B;
        ax(D, Q), Mc(D);
      }
      d.establishStackSpace = yh;
      function Ic(N) {
        if ($)
          return Br(1, 0, N);
        try {
          Sc(N);
        } catch (D) {
          bh(D);
        }
      }
      var Mr = [];
      function Ei(N) {
        var D = Mr[N];
        return D || (N >= Mr.length && (Mr.length = N + 1), Mr[N] = D = Fn.get(N)), D;
      }
      function vh(N, D) {
        return Ei(N)(D);
      }
      d.invokeEntryPoint = vh;
      function Kv() {
        var N = new Error();
        if (!N.stack) {
          try {
            throw new Error();
          } catch (D) {
            N = D;
          }
          if (!N.stack)
            return "(no stack trace available)";
        }
        return N.stack.toString();
      }
      function xh(N, D, B) {
        $e.tlsInitFunctions.push(N);
      }
      function Xv(N, D) {
        Fn.set(N, D), Mr[N] = D;
      }
      var Lr;
      I ? Lr = () => {
        var N = process.hrtime();
        return N[0] * 1e3 + N[1] / 1e6;
      } : $ ? Lr = () => performance.now() - d.__performance_now_clock_drift : Lr = () => performance.now();
      var wh = true;
      function kh(N) {
        return l()[ex() >> 2] = N, N;
      }
      function Sh(N, D) {
        var B;
        if (N === 0)
          B = Date.now();
        else if ((N === 1 || N === 4) && wh)
          B = Lr();
        else
          return kh(28), -1;
        return l()[D >> 2] = B / 1e3 | 0, l()[D + 4 >> 2] = B % 1e3 * 1e3 * 1e3 | 0, 0;
      }
      function Ih(N, D) {
        return Sh(N, D);
      }
      function Ch(N) {
        nx(N, !k, 1, !x), $e.threadInit();
      }
      function Nh(N) {
        $ ? postMessage({ cmd: "cleanupThread", thread: N }) : kc(N);
      }
      function Cc(N) {
        var D = $e.getNewWorker();
        if (!D)
          return 6;
        $e.runningWorkers.push(D);
        var B = $e.pthreads[N.pthread_ptr] = { worker: D, threadInfoStruct: N.pthread_ptr };
        D.pthread = B;
        var Q = { cmd: "run", start_routine: N.startRoutine, arg: N.arg, threadInfoStruct: N.pthread_ptr };
        return D.runPthread = () => {
          Q.time = performance.now(), D.postMessage(Q, N.transferList);
        }, D.loaded && (D.runPthread(), delete D.runPthread), 0;
      }
      function Th(N, D, B, Q) {
        if (typeof SharedArrayBuffer == "undefined")
          return J("Current environment does not support SharedArrayBuffer, pthreads are not available!"), 6;
        var ue = [], pe = 0;
        if ($ && (ue.length === 0 || pe))
          return sx(687865856, N, D, B, Q);
        if (pe)
          return pe;
        var ye = { startRoutine: B, pthread_ptr: N, arg: Q, transferList: ue };
        return $ ? (ye.cmd = "spawnThread", postMessage(ye, ue), 0) : Cc(ye);
      }
      function $h() {
        return 2097152;
      }
      function _h(N, D) {
        if (N == D)
          postMessage({ cmd: "processQueuedMainThreadWork" });
        else if ($)
          postMessage({ targetThread: N, cmd: "processThreadQueue" });
        else {
          var B = $e.pthreads[N], Q = B && B.worker;
          if (!Q)
            return;
          Q.postMessage({ cmd: "processThreadQueue" });
        }
        return 1;
      }
      function Ah() {
        $i("");
      }
      function Eh() {
        I || k || ne("Blocking on the main thread is very dangerous, see https://emscripten.org/docs/porting/pthreads.html#blocking-on-the-main-browser-thread");
      }
      function Nc() {
        return 2147483648;
      }
      function Rh(N, D, B) {
        i().copyWithin(N, D, D + B);
      }
      function Dh() {
        return I ? p$().cpus().length : navigator.hardwareConcurrency;
      }
      function Br(N, D) {
        var B = arguments.length - 2, Q = arguments;
        return Ai(function() {
          for (var ue = B, pe = zi(ue * 8), ye = pe >> 3, Te = 0; Te < B; Te++) {
            var bt = Q[2 + Te];
            p()[ye + Te] = bt;
          }
          return rx(N, ue, pe, D);
        });
      }
      var wu = [];
      function Fh(N, D, B) {
        wu.length = D;
        for (var Q = B >> 3, ue = 0; ue < D; ue++)
          wu[ue] = p()[Q + ue];
        var pe = N < 0, ye = pe ? fh[-N - 1] : Jh[N];
        return ye.apply(null, wu);
      }
      function Oh(N) {
        try {
          return Ce.grow(N - nn.byteLength + 65535 >>> 16), rs(Ce.buffer), 1;
        } catch (D) {
        }
      }
      function Ph(N) {
        var D = i().length;
        if (N = N >>> 0, N <= D)
          return false;
        var B = Nc();
        if (N > B)
          return false;
        for (var Q = 1; Q <= 4; Q *= 2) {
          var ue = D * (1 + 0.2 / Q);
          ue = Math.min(ue, N + 100663296);
          var pe = Math.min(B, Ni(Math.max(N, ue), 65536)), ye = Oh(pe);
          if (ye)
            return true;
        }
        return false;
      }
      var Me = { inEventHandler: 0, removeAllEventListeners: function() {
        for (var N = Me.eventHandlers.length - 1; N >= 0; --N)
          Me._removeHandler(N);
        Me.eventHandlers = [], Me.deferredCalls = [];
      }, registerRemoveEventListeners: function() {
        Me.removeEventListenersRegistered || (uh.push(Me.removeAllEventListeners), Me.removeEventListenersRegistered = true);
      }, deferredCalls: [], deferCall: function(N, D, B) {
        function Q(ye, Te) {
          if (ye.length != Te.length)
            return false;
          for (var bt in ye)
            if (ye[bt] != Te[bt])
              return false;
          return true;
        }
        for (var ue in Me.deferredCalls) {
          var pe = Me.deferredCalls[ue];
          if (pe.targetFunction == N && Q(pe.argsList, B))
            return;
        }
        Me.deferredCalls.push({ targetFunction: N, precedence: D, argsList: B }), Me.deferredCalls.sort(function(ye, Te) {
          return ye.precedence < Te.precedence;
        });
      }, removeDeferredCalls: function(N) {
        for (var D = 0; D < Me.deferredCalls.length; ++D)
          Me.deferredCalls[D].targetFunction == N && (Me.deferredCalls.splice(D, 1), --D);
      }, canPerformEventHandlerRequests: function() {
        return Me.inEventHandler && Me.currentEventHandler.allowsDeferredCalls;
      }, runDeferredCalls: function() {
        if (!!Me.canPerformEventHandlerRequests())
          for (var N = 0; N < Me.deferredCalls.length; ++N) {
            var D = Me.deferredCalls[N];
            Me.deferredCalls.splice(N, 1), --N, D.targetFunction.apply(null, D.argsList);
          }
      }, eventHandlers: [], removeAllHandlersOnTarget: function(N, D) {
        for (var B = 0; B < Me.eventHandlers.length; ++B)
          Me.eventHandlers[B].target == N && (!D || D == Me.eventHandlers[B].eventTypeString) && Me._removeHandler(B--);
      }, _removeHandler: function(N) {
        var D = Me.eventHandlers[N];
        D.target.removeEventListener(D.eventTypeString, D.eventListenerFunc, D.useCapture), Me.eventHandlers.splice(N, 1);
      }, registerOrRemoveHandler: function(N) {
        var D = function(ue) {
          ++Me.inEventHandler, Me.currentEventHandler = N, Me.runDeferredCalls(), N.handlerFunc(ue), Me.runDeferredCalls(), --Me.inEventHandler;
        };
        if (N.callbackfunc)
          N.eventListenerFunc = D, N.target.addEventListener(N.eventTypeString, D, N.useCapture), Me.eventHandlers.push(N), Me.registerRemoveEventListeners();
        else
          for (var B = 0; B < Me.eventHandlers.length; ++B)
            Me.eventHandlers[B].target == N.target && Me.eventHandlers[B].eventTypeString == N.eventTypeString && Me._removeHandler(B--);
      }, queueEventHandlerOnThread_iiii: function(N, D, B, Q, ue) {
        Ai(function() {
          var pe = zi(12);
          l()[pe >> 2] = B, l()[pe + 4 >> 2] = Q, l()[pe + 8 >> 2] = ue, Pf(N, 637534208, D, Q, pe);
        });
      }, getTargetThreadForEventCallback: function(N) {
        switch (N) {
          case 1:
            return 0;
          case 2:
            return $e.currentProxiedOperationCallerThread;
          default:
            return N;
        }
      }, getNodeNameForTarget: function(N) {
        return N ? N == window ? "#window" : N == screen ? "#screen" : N && N.nodeName ? N.nodeName : "" : "";
      }, fullscreenEnabled: function() {
        return document.fullscreenEnabled || document.webkitFullscreenEnabled;
      } };
      function zh(N) {
        var D = Ci(N) + 1, B = Of(D);
        return Ms(N, B, D), B;
      }
      function Mh(N, D, B, Q) {
        Ai(function() {
          var ue = zi(12), pe = 0;
          D && (pe = zh(D)), l()[ue >> 2] = pe, l()[ue + 4 >> 2] = B, l()[ue + 8 >> 2] = Q, Pf(N, 657457152, 0, pe, ue);
        });
      }
      function Lh(N, D, B, Q) {
        D = D ? tn(D) : "", Mh(N, D, B, Q);
      }
      function Bh(N) {
        return N > 2 ? tn(N) : N;
      }
      var Vh = [0, typeof document != "undefined" ? document : 0, typeof window != "undefined" ? window : 0];
      function Wh(N) {
        N = Bh(N);
        var D = Vh[N] || (typeof document != "undefined" ? document.querySelector(N) : void 0);
        return D;
      }
      function ku(N) {
        return Wh(N);
      }
      function Tc(N, D, B) {
        var Q = ku(N);
        if (!Q)
          return -4;
        if (Q.canvasSharedPtr && (l()[Q.canvasSharedPtr >> 2] = D, l()[Q.canvasSharedPtr + 4 >> 2] = B), Q.offscreenCanvas || !Q.controlTransferredOffscreen) {
          Q.offscreenCanvas && (Q = Q.offscreenCanvas);
          var ue = false;
          if (Q.GLctxObject && Q.GLctxObject.GLctx) {
            var pe = Q.GLctxObject.GLctx.getParameter(2978);
            ue = pe[0] === 0 && pe[1] === 0 && pe[2] === Q.width && pe[3] === Q.height;
          }
          Q.width = D, Q.height = B, ue && Q.GLctxObject.GLctx.viewport(0, 0, D, B);
        } else if (Q.canvasSharedPtr) {
          var ye = l()[Q.canvasSharedPtr + 8 >> 2];
          return Lh(ye, N, D, B), 1;
        } else
          return -4;
        return 0;
      }
      function $c(N, D, B) {
        return $ ? Br(2, 1, N, D, B) : Tc(N, D, B);
      }
      function Uh(N, D, B) {
        var Q = ku(N);
        return Q ? Tc(N, D, B) : $c(N, D, B);
      }
      function Gh() {
        throw "unwind";
      }
      function Hh(N) {
        var D = N.getExtension("ANGLE_instanced_arrays");
        if (D)
          return N.vertexAttribDivisor = function(B, Q) {
            D.vertexAttribDivisorANGLE(B, Q);
          }, N.drawArraysInstanced = function(B, Q, ue, pe) {
            D.drawArraysInstancedANGLE(B, Q, ue, pe);
          }, N.drawElementsInstanced = function(B, Q, ue, pe, ye) {
            D.drawElementsInstancedANGLE(B, Q, ue, pe, ye);
          }, 1;
      }
      function qh(N) {
        var D = N.getExtension("OES_vertex_array_object");
        if (D)
          return N.createVertexArray = function() {
            return D.createVertexArrayOES();
          }, N.deleteVertexArray = function(B) {
            D.deleteVertexArrayOES(B);
          }, N.bindVertexArray = function(B) {
            D.bindVertexArrayOES(B);
          }, N.isVertexArray = function(B) {
            return D.isVertexArrayOES(B);
          }, 1;
      }
      function jh(N) {
        var D = N.getExtension("WEBGL_draw_buffers");
        if (D)
          return N.drawBuffers = function(B, Q) {
            D.drawBuffersWEBGL(B, Q);
          }, 1;
      }
      function Kh(N) {
        return !!(N.multiDrawWebgl = N.getExtension("WEBGL_multi_draw"));
      }
      var gt = { counter: 1, buffers: [], programs: [], framebuffers: [], renderbuffers: [], textures: [], shaders: [], vaos: [], contexts: {}, offscreenCanvases: {}, queries: [], stringCache: {}, unpackAlignment: 4, recordError: function(D) {
        gt.lastError || (gt.lastError = D);
      }, getNewId: function(N) {
        for (var D = gt.counter++, B = N.length; B < D; B++)
          N[B] = null;
        return D;
      }, getSource: function(N, D, B, Q) {
        for (var ue = "", pe = 0; pe < D; ++pe) {
          var ye = Q ? l()[Q + pe * 4 >> 2] : -1;
          ue += tn(l()[B + pe * 4 >> 2], ye < 0 ? void 0 : ye);
        }
        return ue;
      }, createContext: function(N, D) {
        N.getContextSafariWebGL2Fixed || (N.getContextSafariWebGL2Fixed = N.getContext, N.getContext = function(ue, pe) {
          var ye = N.getContextSafariWebGL2Fixed(ue, pe);
          return ue == "webgl" == ye instanceof WebGLRenderingContext ? ye : null;
        });
        var B = N.getContext("webgl", D);
        if (!B)
          return 0;
        var Q = gt.registerContext(B, D);
        return Q;
      }, registerContext: function(N, D) {
        var B = Of(8);
        l()[B + 4 >> 2] = zc();
        var Q = { handle: B, attributes: D, version: D.majorVersion, GLctx: N };
        return N.canvas && (N.canvas.GLctxObject = Q), gt.contexts[B] = Q, (typeof D.enableExtensionsByDefault == "undefined" || D.enableExtensionsByDefault) && gt.initExtensions(Q), B;
      }, makeContextCurrent: function(N) {
        return gt.currentContext = gt.contexts[N], d.ctx = Rc = gt.currentContext && gt.currentContext.GLctx, !(N && !Rc);
      }, getContext: function(N) {
        return gt.contexts[N];
      }, deleteContext: function(N) {
        gt.currentContext === gt.contexts[N] && (gt.currentContext = null), typeof Me == "object" && Me.removeAllHandlersOnTarget(gt.contexts[N].GLctx.canvas), gt.contexts[N] && gt.contexts[N].GLctx.canvas && (gt.contexts[N].GLctx.canvas.GLctxObject = void 0), Jv(gt.contexts[N].handle), gt.contexts[N] = null;
      }, initExtensions: function(N) {
        if (N || (N = gt.currentContext), !N.initExtensionsDone) {
          N.initExtensionsDone = true;
          var D = N.GLctx;
          Hh(D), qh(D), jh(D), D.disjointTimerQueryExt = D.getExtension("EXT_disjoint_timer_query"), Kh(D);
          var B = D.getSupportedExtensions() || [];
          B.forEach(function(Q) {
            !Q.includes("lose_context") && !Q.includes("debug") && D.getExtension(Q);
          });
        }
      } }, Xh = ["default", "low-power", "high-performance"];
      function Yh(N, D) {
        var B = D >> 2, Q = l()[B + 6], ue = { alpha: !!l()[B + 0], depth: !!l()[B + 1], stencil: !!l()[B + 2], antialias: !!l()[B + 3], premultipliedAlpha: !!l()[B + 4], preserveDrawingBuffer: !!l()[B + 5], powerPreference: Xh[Q], failIfMajorPerformanceCaveat: !!l()[B + 7], majorVersion: l()[B + 8], minorVersion: l()[B + 9], enableExtensionsByDefault: l()[B + 10], explicitSwapControl: l()[B + 11], proxyContextToMainThread: l()[B + 12], renderViaOffscreenBackBuffer: l()[B + 13] }, pe = ku(N);
        if (!pe || ue.explicitSwapControl)
          return 0;
        var ye = gt.createContext(pe, ue);
        return ye;
      }
      function Qh(N, D) {
        return Yh(N, D);
      }
      var Ri = { mappings: {}, buffers: [null, [], []], printChar: function(N, D) {
        var B = Ri.buffers[N];
        D === 0 || D === 10 ? ((N === 1 ? te : J)(Dn(B, 0)), B.length = 0) : B.push(D);
      }, varargs: void 0, get: function() {
        Ri.varargs += 4;
        var N = l()[Ri.varargs - 4 >> 2];
        return N;
      }, getStr: function(N) {
        var D = tn(N);
        return D;
      }, get64: function(N, D) {
        return N;
      } };
      function _c(N) {
        return $ ? Br(3, 1, N) : 0;
      }
      function Ac(N, D, B, Q, ue) {
        if ($)
          return Br(4, 1, N, D, B, Q, ue);
      }
      function Ec(N, D, B, Q) {
        if ($)
          return Br(5, 1, N, D, B, Q);
        for (var ue = 0, pe = 0; pe < B; pe++) {
          var ye = l()[D >> 2], Te = l()[D + 4 >> 2];
          D += 8;
          for (var bt = 0; bt < Te; bt++)
            Ri.printChar(N, i()[ye + bt]);
          ue += Te;
        }
        return l()[Q >> 2] = ue, 0;
      }
      function Zh(N) {
        Re(N);
      }
      $e.init();
      var Rc, Jh = [null, Ic, $c, _c, Ac, Ec], Yv = false, Dc = { __clock_gettime: Ih, __emscripten_init_main_thread_js: Ch, __emscripten_thread_cleanup: Nh, __pthread_create_js: Th, _emscripten_default_pthread_stack_size: $h, _emscripten_notify_thread_queue: _h, abort: Ah, emscripten_check_blocking_allowed: Eh, emscripten_get_heap_max: Nc, emscripten_get_now: Lr, emscripten_memcpy_big: Rh, emscripten_num_logical_cores: Dh, emscripten_receive_on_main_thread_js: Fh, emscripten_resize_heap: Ph, emscripten_set_canvas_element_size: Uh, emscripten_unwind_to_js_event_loop: Gh, emscripten_webgl_create_context: Qh, exit: Sc, fd_close: _c, fd_seek: Ac, fd_write: Ec, memory: Ce || d.wasmMemory, setTempRet0: Zh }, Qv = hh(), ef = d.___wasm_call_ctors = function() {
        return (ef = d.___wasm_call_ctors = d.asm.__wasm_call_ctors).apply(null, arguments);
      }, tf = d._init = function() {
        return (tf = d._init = d.asm.init).apply(null, arguments);
      }, nf = d._init_with_threads_count = function() {
        return (nf = d._init_with_threads_count = d.asm.init_with_threads_count).apply(null, arguments);
      }, sf = d._get_threads_count = function() {
        return (sf = d._get_threads_count = d.asm.get_threads_count).apply(null, arguments);
      }, rf = d._register_tensor = function() {
        return (rf = d._register_tensor = d.asm.register_tensor).apply(null, arguments);
      }, af = d._dispose_data = function() {
        return (af = d._dispose_data = d.asm.dispose_data).apply(null, arguments);
      }, of = d._dispose = function() {
        return (of = d._dispose = d.asm.dispose).apply(null, arguments);
      }, uf = d._Abs = function() {
        return (uf = d._Abs = d.asm.Abs).apply(null, arguments);
      }, lf = d._Add = function() {
        return (lf = d._Add = d.asm.Add).apply(null, arguments);
      }, cf = d._AddN = function() {
        return (cf = d._AddN = d.asm.AddN).apply(null, arguments);
      }, df = d._All = function() {
        return (df = d._All = d.asm.All).apply(null, arguments);
      }, pf = d._Any = function() {
        return (pf = d._Any = d.asm.Any).apply(null, arguments);
      }, hf = d._ArgMax = function() {
        return (hf = d._ArgMax = d.asm.ArgMax).apply(null, arguments);
      }, ff = d._AvgPool = function() {
        return (ff = d._AvgPool = d.asm.AvgPool).apply(null, arguments);
      }, mf = d._BatchMatMul = function() {
        return (mf = d._BatchMatMul = d.asm.BatchMatMul).apply(null, arguments);
      }, gf = d._Ceil = function() {
        return (gf = d._Ceil = d.asm.Ceil).apply(null, arguments);
      }, bf = d._ClipByValue = function() {
        return (bf = d._ClipByValue = d.asm.ClipByValue).apply(null, arguments);
      }, yf = d._Conv2D = function() {
        return (yf = d._Conv2D = d.asm.Conv2D).apply(null, arguments);
      }, vf = d._Conv2DBackpropInput = function() {
        return (vf = d._Conv2DBackpropInput = d.asm.Conv2DBackpropInput).apply(null, arguments);
      }, xf = d._Cos = function() {
        return (xf = d._Cos = d.asm.Cos).apply(null, arguments);
      }, wf = d._Cosh = function() {
        return (wf = d._Cosh = d.asm.Cosh).apply(null, arguments);
      }, kf = d._CropAndResize = function() {
        return (kf = d._CropAndResize = d.asm.CropAndResize).apply(null, arguments);
      }, Sf = d._Cumprod = function() {
        return (Sf = d._Cumprod = d.asm.Cumprod).apply(null, arguments);
      }, If = d._Cumsum = function() {
        return (If = d._Cumsum = d.asm.Cumsum).apply(null, arguments);
      }, Cf = d._DepthToSpace = function() {
        return (Cf = d._DepthToSpace = d.asm.DepthToSpace).apply(null, arguments);
      }, Nf = d._DepthwiseConv2dNative = function() {
        return (Nf = d._DepthwiseConv2dNative = d.asm.DepthwiseConv2dNative).apply(null, arguments);
      }, Tf = d._Elu = function() {
        return (Tf = d._Elu = d.asm.Elu).apply(null, arguments);
      }, $f = d._Equal = function() {
        return ($f = d._Equal = d.asm.Equal).apply(null, arguments);
      }, _f = d._Exp = function() {
        return (_f = d._Exp = d.asm.Exp).apply(null, arguments);
      }, Af = d._FlipLeftRight = function() {
        return (Af = d._FlipLeftRight = d.asm.FlipLeftRight).apply(null, arguments);
      }, Fc = d._Floor = function() {
        return (Fc = d._Floor = d.asm.Floor).apply(null, arguments);
      }, Oc = d._FloorDiv = function() {
        return (Oc = d._FloorDiv = d.asm.FloorDiv).apply(null, arguments);
      }, Su = d._FusedBatchNorm = function() {
        return (Su = d._FusedBatchNorm = d.asm.FusedBatchNorm).apply(null, arguments);
      }, Ef = d._FusedConv2D = function() {
        return (Ef = d._FusedConv2D = d.asm.FusedConv2D).apply(null, arguments);
      }, Rf = d._FusedDepthwiseConv2D = function() {
        return (Rf = d._FusedDepthwiseConv2D = d.asm.FusedDepthwiseConv2D).apply(null, arguments);
      }, Di = d._Gather = function() {
        return (Di = d._Gather = d.asm.Gather).apply(null, arguments);
      }, Iu = d._GatherNd = function() {
        return (Iu = d._GatherNd = d.asm.GatherNd).apply(null, arguments);
      }, Cu = d._Greater = function() {
        return (Cu = d._Greater = d.asm.Greater).apply(null, arguments);
      }, Zv = d._GreaterEqual = function() {
        return (Zv = d._GreaterEqual = d.asm.GreaterEqual).apply(null, arguments);
      }, Fi = d._LeakyRelu = function() {
        return (Fi = d._LeakyRelu = d.asm.LeakyRelu).apply(null, arguments);
      }, Oi = d._Less = function() {
        return (Oi = d._Less = d.asm.Less).apply(null, arguments);
      }, Df = d._LessEqual = function() {
        return (Df = d._LessEqual = d.asm.LessEqual).apply(null, arguments);
      }, H = d._Log = function() {
        return (H = d._Log = d.asm.Log).apply(null, arguments);
      }, ee = d._LogicalAnd = function() {
        return (ee = d._LogicalAnd = d.asm.LogicalAnd).apply(null, arguments);
      }, ce = d._Max = function() {
        return (ce = d._Max = d.asm.Max).apply(null, arguments);
      }, Se = d._MaxPool = function() {
        return (Se = d._MaxPool = d.asm.MaxPool).apply(null, arguments);
      }, Qe = d._Maximum = function() {
        return (Qe = d._Maximum = d.asm.Maximum).apply(null, arguments);
      }, Ze = d._Mean = function() {
        return (Ze = d._Mean = d.asm.Mean).apply(null, arguments);
      }, Le = d._Min = function() {
        return (Le = d._Min = d.asm.Min).apply(null, arguments);
      }, ze = d._Minimum = function() {
        return (ze = d._Minimum = d.asm.Minimum).apply(null, arguments);
      }, Tt = d._MirrorPad = function() {
        return (Tt = d._MirrorPad = d.asm.MirrorPad).apply(null, arguments);
      }, is = d._Multiply = function() {
        return (is = d._Multiply = d.asm.Multiply).apply(null, arguments);
      }, os = d._Neg = function() {
        return (os = d._Neg = d.asm.Neg).apply(null, arguments);
      }, Pi = d._NonMaxSuppressionV3 = function() {
        return (Pi = d._NonMaxSuppressionV3 = d.asm.NonMaxSuppressionV3).apply(null, arguments);
      }, Vr = d._NonMaxSuppressionV4 = function() {
        return (Vr = d._NonMaxSuppressionV4 = d.asm.NonMaxSuppressionV4).apply(null, arguments);
      }, Ff = d._NonMaxSuppressionV5 = function() {
        return (Ff = d._NonMaxSuppressionV5 = d.asm.NonMaxSuppressionV5).apply(null, arguments);
      }, an = d._NotEqual = function() {
        return (an = d._NotEqual = d.asm.NotEqual).apply(null, arguments);
      }, tr = d._OneHot = function() {
        return (tr = d._OneHot = d.asm.OneHot).apply(null, arguments);
      }, Pc = d._PadV2 = function() {
        return (Pc = d._PadV2 = d.asm.PadV2).apply(null, arguments);
      }, tT = d._Pow = function() {
        return (tT = d._Pow = d.asm.Pow).apply(null, arguments);
      }, nT = d._Prelu = function() {
        return (nT = d._Prelu = d.asm.Prelu).apply(null, arguments);
      }, sT = d._Prod = function() {
        return (sT = d._Prod = d.asm.Prod).apply(null, arguments);
      }, rT = d._RealDiv = function() {
        return (rT = d._RealDiv = d.asm.RealDiv).apply(null, arguments);
      }, aT = d._Relu = function() {
        return (aT = d._Relu = d.asm.Relu).apply(null, arguments);
      }, iT = d._Relu6 = function() {
        return (iT = d._Relu6 = d.asm.Relu6).apply(null, arguments);
      }, oT = d._ResizeBilinear = function() {
        return (oT = d._ResizeBilinear = d.asm.ResizeBilinear).apply(null, arguments);
      }, uT = d._Reverse = function() {
        return (uT = d._Reverse = d.asm.Reverse).apply(null, arguments);
      }, lT = d._RotateWithOffset = function() {
        return (lT = d._RotateWithOffset = d.asm.RotateWithOffset).apply(null, arguments);
      }, cT = d._Round = function() {
        return (cT = d._Round = d.asm.Round).apply(null, arguments);
      }, dT = d._Rsqrt = function() {
        return (dT = d._Rsqrt = d.asm.Rsqrt).apply(null, arguments);
      }, pT = d._ScatterNd = function() {
        return (pT = d._ScatterNd = d.asm.ScatterNd).apply(null, arguments);
      }, hT = d._SelectV2 = function() {
        return (hT = d._SelectV2 = d.asm.SelectV2).apply(null, arguments);
      }, fT = d._Sigmoid = function() {
        return (fT = d._Sigmoid = d.asm.Sigmoid).apply(null, arguments);
      }, mT = d._Sin = function() {
        return (mT = d._Sin = d.asm.Sin).apply(null, arguments);
      }, gT = d._Softmax = function() {
        return (gT = d._Softmax = d.asm.Softmax).apply(null, arguments);
      }, bT = d._SparseFillEmptyRows = function() {
        return (bT = d._SparseFillEmptyRows = d.asm.SparseFillEmptyRows).apply(null, arguments);
      }, yT = d._SparseReshape = function() {
        return (yT = d._SparseReshape = d.asm.SparseReshape).apply(null, arguments);
      }, vT = d._SparseSegmentReduction = function() {
        return (vT = d._SparseSegmentReduction = d.asm.SparseSegmentReduction).apply(null, arguments);
      }, xT = d._Sqrt = function() {
        return (xT = d._Sqrt = d.asm.Sqrt).apply(null, arguments);
      }, wT = d._Square = function() {
        return (wT = d._Square = d.asm.Square).apply(null, arguments);
      }, kT = d._SquaredDifference = function() {
        return (kT = d._SquaredDifference = d.asm.SquaredDifference).apply(null, arguments);
      }, ST = d._Step = function() {
        return (ST = d._Step = d.asm.Step).apply(null, arguments);
      }, IT = d._StridedSlice = function() {
        return (IT = d._StridedSlice = d.asm.StridedSlice).apply(null, arguments);
      }, CT = d._Sub = function() {
        return (CT = d._Sub = d.asm.Sub).apply(null, arguments);
      }, NT = d._Sum = function() {
        return (NT = d._Sum = d.asm.Sum).apply(null, arguments);
      }, TT = d._Tan = function() {
        return (TT = d._Tan = d.asm.Tan).apply(null, arguments);
      }, $T = d._Tanh = function() {
        return ($T = d._Tanh = d.asm.Tanh).apply(null, arguments);
      }, _T = d._Tile = function() {
        return (_T = d._Tile = d.asm.Tile).apply(null, arguments);
      }, AT = d._TopK = function() {
        return (AT = d._TopK = d.asm.TopK).apply(null, arguments);
      }, ET = d._Transform = function() {
        return (ET = d._Transform = d.asm.Transform).apply(null, arguments);
      }, RT = d._Transpose = function() {
        return (RT = d._Transpose = d.asm.Transpose).apply(null, arguments);
      }, DT = d.__FusedMatMul = function() {
        return (DT = d.__FusedMatMul = d.asm._FusedMatMul).apply(null, arguments);
      }, Of = d._malloc = function() {
        return (Of = d._malloc = d.asm.malloc).apply(null, arguments);
      }, Jv = d._free = function() {
        return (Jv = d._free = d.asm.free).apply(null, arguments);
      }, FT = d._emscripten_tls_init = function() {
        return (FT = d._emscripten_tls_init = d.asm.emscripten_tls_init).apply(null, arguments);
      }, ex = d.___errno_location = function() {
        return (ex = d.___errno_location = d.asm.__errno_location).apply(null, arguments);
      }, zc = d._pthread_self = function() {
        return (zc = d._pthread_self = d.asm.pthread_self).apply(null, arguments);
      }, tx = d._emscripten_main_thread_process_queued_calls = function() {
        return (tx = d._emscripten_main_thread_process_queued_calls = d.asm.emscripten_main_thread_process_queued_calls).apply(null, arguments);
      }, OT = d.__emscripten_thread_crashed = function() {
        return (OT = d.__emscripten_thread_crashed = d.asm._emscripten_thread_crashed).apply(null, arguments);
      }, nx = d.__emscripten_thread_init = function() {
        return (nx = d.__emscripten_thread_init = d.asm._emscripten_thread_init).apply(null, arguments);
      }, PT = d._emscripten_current_thread_process_queued_calls = function() {
        return (PT = d._emscripten_current_thread_process_queued_calls = d.asm.emscripten_current_thread_process_queued_calls).apply(null, arguments);
      }, zT = d._emscripten_main_browser_thread_id = function() {
        return (zT = d._emscripten_main_browser_thread_id = d.asm.emscripten_main_browser_thread_id).apply(null, arguments);
      }, MT = d._emscripten_sync_run_in_main_thread_2 = function() {
        return (MT = d._emscripten_sync_run_in_main_thread_2 = d.asm.emscripten_sync_run_in_main_thread_2).apply(null, arguments);
      }, sx = d._emscripten_sync_run_in_main_thread_4 = function() {
        return (sx = d._emscripten_sync_run_in_main_thread_4 = d.asm.emscripten_sync_run_in_main_thread_4).apply(null, arguments);
      }, rx = d._emscripten_run_in_main_runtime_thread_js = function() {
        return (rx = d._emscripten_run_in_main_runtime_thread_js = d.asm.emscripten_run_in_main_runtime_thread_js).apply(null, arguments);
      }, Pf = d._emscripten_dispatch_to_thread_ = function() {
        return (Pf = d._emscripten_dispatch_to_thread_ = d.asm.emscripten_dispatch_to_thread_).apply(null, arguments);
      }, zf = d.__emscripten_thread_free_data = function() {
        return (zf = d.__emscripten_thread_free_data = d.asm._emscripten_thread_free_data).apply(null, arguments);
      }, LT = d.__emscripten_thread_exit = function() {
        return (LT = d.__emscripten_thread_exit = d.asm._emscripten_thread_exit).apply(null, arguments);
      }, BT = d._memalign = function() {
        return (BT = d._memalign = d.asm.memalign).apply(null, arguments);
      }, ax = d._emscripten_stack_set_limits = function() {
        return (ax = d._emscripten_stack_set_limits = d.asm.emscripten_stack_set_limits).apply(null, arguments);
      }, Mf = d.stackSave = function() {
        return (Mf = d.stackSave = d.asm.stackSave).apply(null, arguments);
      }, Mc = d.stackRestore = function() {
        return (Mc = d.stackRestore = d.asm.stackRestore).apply(null, arguments);
      }, zi = d.stackAlloc = function() {
        return (zi = d.stackAlloc = d.asm.stackAlloc).apply(null, arguments);
      }, VT = d.dynCall_iijjiiii = function() {
        return (VT = d.dynCall_iijjiiii = d.asm.dynCall_iijjiiii).apply(null, arguments);
      }, WT = d.dynCall_jiji = function() {
        return (WT = d.dynCall_jiji = d.asm.dynCall_jiji).apply(null, arguments);
      }, ix = d.__emscripten_allow_main_runtime_queued_calls = 21456;
      d.cwrap = en, d.keepRuntimeAlive = zr, d.PThread = $e, d.PThread = $e, d.wasmMemory = Ce, d.ExitStatus = Nu;
      var Lc;
      function Nu(N) {
        this.name = "ExitStatus", this.message = "Program terminated with exit(" + N + ")", this.status = N;
      }
      as = function N() {
        Lc || Lf(), Lc || (as = N);
      };
      function Lf(N) {
        if (N = N || b, er > 0)
          return;
        if ($) {
          h(d), yu(), postMessage({ cmd: "loaded" });
          return;
        }
        if (sn(), er > 0)
          return;
        function D() {
          Lc || (Lc = true, d.calledRun = true, !at && (yu(), h(d), d.onRuntimeInitialized && d.onRuntimeInitialized(), dh()));
        }
        d.setStatus ? (d.setStatus("Running..."), setTimeout(function() {
          setTimeout(function() {
            d.setStatus("");
          }, 1), D();
        }, 1)) : D();
      }
      d.run = Lf;
      function UT(N, D) {
        if (Jt = N, !D && $)
          throw Ic(N), "unwind";
        zr() || ch(), GT(N);
      }
      function GT(N) {
        Jt = N, zr() || ($e.terminateAllThreads(), d.onExit && d.onExit(N), at = true), v(N, new Nu(N));
      }
      if (d.preInit)
        for (typeof d.preInit == "function" && (d.preInit = [d.preInit]); d.preInit.length > 0; )
          d.preInit.pop()();
      Lf();
      var Bc;
      m && (Bc = { uncaughtException: process.listeners("uncaughtException").filter(function(N) {
        return !m.uncaughtException.indexOf(N) > -1;
      }), unhandledRejection: process.listeners("unhandledRejection").filter(function(N) {
        return !m.unhandledRejection.indexOf(N) > -1;
      }) });
      var Vc;
      if (typeof WasmBackendModule != "undefined")
        Vc = WasmBackendModule;
      else if (typeof r != "undefined")
        Vc = r;
      else
        throw new Error("Could not find wasm module in post.js");
      if (Bc) {
        var HT = Vc._dispose;
        Vc._dispose = function() {
          HT(), Bc.uncaughtException.forEach(function(N) {
            process.removeListener("uncaughtException", N);
          }), Bc.unhandledRejection.forEach(function(N) {
            process.removeListener("unhandledRejection", N);
          });
        };
      }
      return r.ready;
    };
  })();
  typeof e == "object" && typeof t == "object" ? t.exports = n : typeof define == "function" && define.amd ? define([], function() {
    return n;
  }) : typeof e == "object" && (e.WasmBackendModuleThreadedSimd = n);
} });
var f$ = Mt({ "src/tfjs-backend-wasm/wasm-out/tfjs-backend-wasm.js"(e, t) {
  var n = (() => {
    var s = typeof document != "undefined" && document.currentScript ? document.currentScript.src : void 0;
    return typeof __filename != "undefined" && (s = s || __filename), function(r) {
      r = r || {};
      var a = typeof r != "undefined" ? r : {}, i, o;
      a.ready = new Promise(function(H, ee) {
        i = H, o = ee;
      });
      var u;
      typeof process != "undefined" && process.listeners && (u = { uncaughtException: process.listeners("uncaughtException"), unhandledRejection: process.listeners("unhandledRejection") });
      var l = Object.assign({}, a), c = [], p = "./this.program", d = (H, ee) => {
        throw ee;
      }, h = typeof window == "object", f = typeof importScripts == "function", m = typeof process == "object" && typeof process.versions == "object" && typeof process.versions.node == "string", g = "";
      function b(H) {
        return a.locateFile ? a.locateFile(H, g) : g + H;
      }
      var y, v, x, k;
      function I(H) {
        if (H instanceof Iu)
          return;
        A("exiting due to exception: " + H);
      }
      var $, R, E;
      m ? (f ? g = md().dirname(g) + "/" : g = __dirname + "/", E = () => {
        R || ($ = og(), R = md());
      }, y = function(ee, ce) {
        return E(), ee = R.normalize(ee), $.readFileSync(ee, ce ? void 0 : "utf8");
      }, x = (H) => {
        var ee = y(H, true);
        return ee.buffer || (ee = new Uint8Array(ee)), ee;
      }, v = (H, ee, ce) => {
        E(), H = R.normalize(H), $.readFile(H, function(Se, Qe) {
          Se ? ce(Se) : ee(Qe.buffer);
        });
      }, process.argv.length > 1 && (p = process.argv[1].replace(/\\/g, "/")), c = process.argv.slice(2), process.on("uncaughtException", function(H) {
        if (!(H instanceof Iu))
          throw H;
      }), process.on("unhandledRejection", function(H) {
        throw H;
      }), d = (H, ee) => {
        if (bu())
          throw process.exitCode = H, ee;
        I(ee), process.exit(H);
      }, a.inspect = function() {
        return "[Emscripten Module object]";
      }) : (h || f) && (f ? g = self.location.href : typeof document != "undefined" && document.currentScript && (g = document.currentScript.src), s && (g = s), g.indexOf("blob:") !== 0 ? g = g.substr(0, g.replace(/[?#].*/, "").lastIndexOf("/") + 1) : g = "", y = (H) => {
        var ee = new XMLHttpRequest();
        return ee.open("GET", H, false), ee.send(null), ee.responseText;
      }, f && (x = (H) => {
        var ee = new XMLHttpRequest();
        return ee.open("GET", H, false), ee.responseType = "arraybuffer", ee.send(null), new Uint8Array(ee.response);
      }), v = (H, ee, ce) => {
        var Se = new XMLHttpRequest();
        Se.open("GET", H, true), Se.responseType = "arraybuffer", Se.onload = () => {
          if (Se.status == 200 || Se.status == 0 && Se.response) {
            ee(Se.response);
            return;
          }
          ce();
        }, Se.onerror = ce, Se.send(null);
      }, k = (H) => document.title = H);
      var P = a.print || console.log.bind(console), A = a.printErr || console.warn.bind(console);
      Object.assign(a, l), l = null, a.arguments && (c = a.arguments), a.thisProgram && (p = a.thisProgram), a.quit && (d = a.quit);
      var O = 4;
      function T(H) {
        T.shown || (T.shown = {}), T.shown[H] || (T.shown[H] = 1, A(H));
      }
      function z(H, ee) {
        if (typeof WebAssembly.Function == "function") {
          for (var ce = { i: "i32", j: "i64", f: "f32", d: "f64" }, Se = { parameters: [], results: ee[0] == "v" ? [] : [ce[ee[0]]] }, Qe = 1; Qe < ee.length; ++Qe)
            Se.parameters.push(ce[ee[Qe]]);
          return new WebAssembly.Function(Se, H);
        }
        var Ze = [1, 0, 1, 96], Le = ee.slice(0, 1), ze = ee.slice(1), Tt = { i: 127, j: 126, f: 125, d: 124 };
        Ze.push(ze.length);
        for (var Qe = 0; Qe < ze.length; ++Qe)
          Ze.push(Tt[ze[Qe]]);
        Le == "v" ? Ze.push(0) : Ze = Ze.concat([1, Tt[Le]]), Ze[1] = Ze.length - 2;
        var is = new Uint8Array([0, 97, 115, 109, 1, 0, 0, 0].concat(Ze, [2, 7, 1, 1, 101, 1, 102, 0, 0, 7, 5, 1, 1, 102, 0, 0])), os = new WebAssembly.Module(is), Pi = new WebAssembly.Instance(os, { e: { f: H } }), Vr = Pi.exports.f;
        return Vr;
      }
      var W = [], q;
      function X() {
        if (W.length)
          return W.pop();
        try {
          Zs.grow(1);
        } catch (H) {
          throw H instanceof RangeError ? "Unable to grow wasm table. Set ALLOW_TABLE_GROWTH." : H;
        }
        return Zs.length - 1;
      }
      function Y(H, ee) {
        for (var ce = H; ce < H + ee; ce++) {
          var Se = xu(ce);
          Se && q.set(Se, ce);
        }
      }
      var Z = 0, te = (H) => {
        Z = H;
      }, J;
      a.wasmBinary && (J = a.wasmBinary);
      var se = a.noExitRuntime || true;
      typeof WebAssembly != "object" && Pr("no native wasm support detected");
      var ne, oe = false, ae;
      function de(H, ee) {
        H || Pr(ee);
      }
      function me(H) {
        var ee = a["_" + H];
        return ee;
      }
      function ke(H, ee, ce, Se, Qe) {
        var Ze = { string: function(an) {
          var tr = 0;
          if (an != null && an !== 0) {
            var Pc = (an.length << 2) + 1;
            tr = Su(Pc), tt(an, tr, Pc);
          }
          return tr;
        }, array: function(an) {
          var tr = Su(an.length);
          return at(an, tr), tr;
        } };
        function Le(an) {
          return ee === "string" ? Je(an) : ee === "boolean" ? Boolean(an) : an;
        }
        var ze = me(H), Tt = [], is = 0;
        if (Se)
          for (var os = 0; os < Se.length; os++) {
            var Pi = Ze[ce[os]];
            Pi ? (is === 0 && (is = Fc()), Tt[os] = Pi(Se[os])) : Tt[os] = Se[os];
          }
        var Vr = ze.apply(null, Tt);
        function Ff(an) {
          return is !== 0 && Oc(is), Le(an);
        }
        return Vr = Ff(Vr), Vr;
      }
      function Ie(H, ee, ce, Se) {
        ce = ce || [];
        var Qe = ce.every(function(Le) {
          return Le === "number";
        }), Ze = ee !== "string";
        return Ze && Qe && !Se ? me(H) : function() {
          return ke(H, ee, ce, arguments, Se);
        };
      }
      var Re = 1, Pe = typeof TextDecoder != "undefined" ? new TextDecoder("utf8") : void 0;
      function Xe(H, ee, ce) {
        for (var Se = ee + ce, Qe = ee; H[Qe] && !(Qe >= Se); )
          ++Qe;
        if (Qe - ee > 16 && H.subarray && Pe)
          return Pe.decode(H.subarray(ee, Qe));
        for (var Ze = ""; ee < Qe; ) {
          var Le = H[ee++];
          if (!(Le & 128)) {
            Ze += String.fromCharCode(Le);
            continue;
          }
          var ze = H[ee++] & 63;
          if ((Le & 224) == 192) {
            Ze += String.fromCharCode((Le & 31) << 6 | ze);
            continue;
          }
          var Tt = H[ee++] & 63;
          if ((Le & 240) == 224 ? Le = (Le & 15) << 12 | ze << 6 | Tt : Le = (Le & 7) << 18 | ze << 12 | Tt << 6 | H[ee++] & 63, Le < 65536)
            Ze += String.fromCharCode(Le);
          else {
            var is = Le - 65536;
            Ze += String.fromCharCode(55296 | is >> 10, 56320 | is & 1023);
          }
        }
        return Ze;
      }
      function Je(H, ee) {
        return H ? Xe(en, H, ee) : "";
      }
      function Ye(H, ee, ce, Se) {
        if (!(Se > 0))
          return 0;
        for (var Qe = ce, Ze = ce + Se - 1, Le = 0; Le < H.length; ++Le) {
          var ze = H.charCodeAt(Le);
          if (ze >= 55296 && ze <= 57343) {
            var Tt = H.charCodeAt(++Le);
            ze = 65536 + ((ze & 1023) << 10) | Tt & 1023;
          }
          if (ze <= 127) {
            if (ce >= Ze)
              break;
            ee[ce++] = ze;
          } else if (ze <= 2047) {
            if (ce + 1 >= Ze)
              break;
            ee[ce++] = 192 | ze >> 6, ee[ce++] = 128 | ze & 63;
          } else if (ze <= 65535) {
            if (ce + 2 >= Ze)
              break;
            ee[ce++] = 224 | ze >> 12, ee[ce++] = 128 | ze >> 6 & 63, ee[ce++] = 128 | ze & 63;
          } else {
            if (ce + 3 >= Ze)
              break;
            ee[ce++] = 240 | ze >> 18, ee[ce++] = 128 | ze >> 12 & 63, ee[ce++] = 128 | ze >> 6 & 63, ee[ce++] = 128 | ze & 63;
          }
        }
        return ee[ce] = 0, ce - Qe;
      }
      function tt(H, ee, ce) {
        return Ye(H, en, ee, ce);
      }
      function Ce(H) {
        for (var ee = 0, ce = 0; ce < H.length; ++ce) {
          var Se = H.charCodeAt(ce);
          Se >= 55296 && Se <= 57343 && (Se = 65536 + ((Se & 1023) << 10) | H.charCodeAt(++ce) & 1023), Se <= 127 ? ++ee : Se <= 2047 ? ee += 2 : Se <= 65535 ? ee += 3 : ee += 4;
        }
        return ee;
      }
      var ut = typeof TextDecoder != "undefined" ? new TextDecoder("utf-16le") : void 0;
      function at(H, ee) {
        Rt.set(H, ee);
      }
      function Jt(H, ee, ce) {
        for (var Se = 0; Se < H.length; ++Se)
          Rt[ee++ >> 0] = H.charCodeAt(Se);
        ce || (Rt[ee >> 0] = 0);
      }
      function Nt(H, ee) {
        return H % ee > 0 && (H += ee - H % ee), H;
      }
      var In, Rt, en, Cn, Nn, Yt, Dn, tn, zs;
      function Ms(H) {
        In = H, a.HEAP8 = Rt = new Int8Array(H), a.HEAP16 = Cn = new Int16Array(H), a.HEAP32 = Yt = new Int32Array(H), a.HEAPU8 = en = new Uint8Array(H), a.HEAPU16 = Nn = new Uint16Array(H), a.HEAPU32 = Dn = new Uint32Array(H), a.HEAPF32 = tn = new Float32Array(H), a.HEAPF64 = zs = new Float64Array(H);
      }
      var Ci = a.INITIAL_MEMORY || 16777216, Zs, Ls = [], gu = [], Ni = [], nn = false, uc = false, lc = 0;
      function bu() {
        return se || lc > 0;
      }
      function cc() {
        if (a.preRun)
          for (typeof a.preRun == "function" && (a.preRun = [a.preRun]); a.preRun.length; )
            hc(a.preRun.shift());
        vu(Ls);
      }
      function dc() {
        nn = true, vu(gu);
      }
      function Vv() {
        uc = true;
      }
      function pc() {
        if (a.postRun)
          for (typeof a.postRun == "function" && (a.postRun = [a.postRun]); a.postRun.length; )
            fc(a.postRun.shift());
        vu(Ni);
      }
      function hc(H) {
        Ls.unshift(H);
      }
      function rs(H) {
        gu.unshift(H);
      }
      function fc(H) {
        Ni.unshift(H);
      }
      var Fn = 0, Ti = null, Js = null;
      function uh(H) {
        Fn++, a.monitorRunDependencies && a.monitorRunDependencies(Fn);
      }
      function mc(H) {
        if (Fn--, a.monitorRunDependencies && a.monitorRunDependencies(Fn), Fn == 0 && (Ti !== null && (clearInterval(Ti), Ti = null), Js)) {
          var ee = Js;
          Js = null, ee();
        }
      }
      a.preloadedImages = {}, a.preloadedAudios = {};
      function Pr(H) {
        a.onAbort && a.onAbort(H), H = "Aborted(" + H + ")", A(H), oe = true, ae = 1, H += ". Build with -s ASSERTIONS=1 for more info.";
        var ee = new WebAssembly.RuntimeError(H);
        throw o(ee), ee;
      }
      var lh = "data:application/octet-stream;base64,";
      function gc(H) {
        return H.startsWith(lh);
      }
      function zr(H) {
        return H.startsWith("file://");
      }
      var sn;
      sn = "tfjs-backend-wasm.wasm", gc(sn) || (sn = b(sn));
      function yu(H) {
        try {
          if (H == sn && J)
            return new Uint8Array(J);
          if (x)
            return x(H);
          throw "both async and sync fetching of the wasm failed";
        } catch (ee) {
          Pr(ee);
        }
      }
      function ch() {
        if (!J && (h || f)) {
          if (typeof fetch == "function" && !zr(sn))
            return fetch(sn, { credentials: "same-origin" }).then(function(H) {
              if (!H.ok)
                throw "failed to load wasm binary file at '" + sn + "'";
              return H.arrayBuffer();
            }).catch(function() {
              return yu(sn);
            });
          if (v)
            return new Promise(function(H, ee) {
              v(sn, function(ce) {
                H(new Uint8Array(ce));
              }, ee);
            });
        }
        return Promise.resolve().then(function() {
          return yu(sn);
        });
      }
      function dh() {
        var H = { env: Ai, wasi_snapshot_preview1: Ai };
        function ee(Le, ze) {
          var Tt = Le.exports;
          a.asm = Tt, ne = a.asm.memory, Ms(ne.buffer), Zs = a.asm.__indirect_function_table, rs(a.asm.__wasm_call_ctors), mc("wasm-instantiate");
        }
        uh("wasm-instantiate");
        function ce(Le) {
          ee(Le.instance);
        }
        function Se(Le) {
          return ch().then(function(ze) {
            return WebAssembly.instantiate(ze, H);
          }).then(function(ze) {
            return ze;
          }).then(Le, function(ze) {
            A("failed to asynchronously prepare wasm: " + ze), Pr(ze);
          });
        }
        function Qe() {
          return !J && typeof WebAssembly.instantiateStreaming == "function" && !gc(sn) && !zr(sn) && typeof fetch == "function" ? fetch(sn, { credentials: "same-origin" }).then(function(Le) {
            var ze = WebAssembly.instantiateStreaming(Le, H);
            return ze.then(ce, function(Tt) {
              return A("wasm streaming compile failed: " + Tt), A("falling back to ArrayBuffer instantiation"), Se(ce);
            });
          }) : Se(ce);
        }
        if (a.instantiateWasm)
          try {
            var Ze = a.instantiateWasm(H, ee);
            return Ze;
          } catch (Le) {
            return A("Module.instantiateWasm callback failed with error: " + Le), false;
          }
        return Qe().catch(o), {};
      }
      var Wv, Uv;
      function vu(H) {
        for (; H.length > 0; ) {
          var ee = H.shift();
          if (typeof ee == "function") {
            ee(a);
            continue;
          }
          var ce = ee.func;
          typeof ce == "number" ? ee.arg === void 0 ? xu(ce)() : xu(ce)(ee.arg) : ce(ee.arg === void 0 ? null : ee.arg);
        }
      }
      function er(H) {
        return H;
      }
      function bc(H) {
        var ee = /\b_Z[\w\d_]+/g;
        return H.replace(ee, function(ce) {
          var Se = ce;
          return ce === Se ? ce : Se + " [" + ce + "]";
        });
      }
      var as = [];
      function xu(H) {
        var ee = as[H];
        return ee || (H >= as.length && (as.length = H + 1), as[H] = ee = Zs.get(H)), ee;
      }
      function Gv() {
        var H = new Error();
        if (!H.stack) {
          try {
            throw new Error();
          } catch (ee) {
            H = ee;
          }
          if (!H.stack)
            return "(no stack trace available)";
        }
        return H.stack.toString();
      }
      function $i(H, ee) {
        Zs.set(H, ee), as[H] = ee;
      }
      function ph() {
        Pr("");
      }
      function yc(H, ee, ce) {
        en.copyWithin(H, ee, ee + ce);
      }
      function vc() {
        return 2147483648;
      }
      function rn(H) {
        try {
          return ne.grow(H - In.byteLength + 65535 >>> 16), Ms(ne.buffer), 1;
        } catch (ee) {
        }
      }
      function xc(H) {
        var ee = en.length;
        H = H >>> 0;
        var ce = vc();
        if (H > ce)
          return false;
        for (var Se = 1; Se <= 4; Se *= 2) {
          var Qe = ee * (1 + 0.2 / Se);
          Qe = Math.min(Qe, H + 100663296);
          var Ze = Math.min(ce, Nt(Math.max(H, Qe), 65536)), Le = rn(Ze);
          if (Le)
            return true;
        }
        return false;
      }
      var _i = { mappings: {}, buffers: [null, [], []], printChar: function(H, ee) {
        var ce = _i.buffers[H];
        ee === 0 || ee === 10 ? ((H === 1 ? P : A)(Xe(ce, 0)), ce.length = 0) : ce.push(ee);
      }, varargs: void 0, get: function() {
        _i.varargs += 4;
        var H = Yt[_i.varargs - 4 >> 2];
        return H;
      }, getStr: function(H) {
        var ee = Je(H);
        return ee;
      }, get64: function(H, ee) {
        return H;
      } };
      function hh(H) {
        return 0;
      }
      function Hv(H, ee, ce, Se, Qe) {
      }
      function qv(H, ee, ce, Se) {
        for (var Qe = 0, Ze = 0; Ze < ce; Ze++) {
          var Le = Yt[ee >> 2], ze = Yt[ee + 4 >> 2];
          ee += 8;
          for (var Tt = 0; Tt < ze; Tt++)
            _i.printChar(H, en[Le + Tt]);
          Qe += ze;
        }
        return Yt[Se >> 2] = Qe, 0;
      }
      function fh(H) {
        te(H);
      }
      var wc = false, Ai = { abort: ph, emscripten_memcpy_big: yc, emscripten_resize_heap: xc, fd_close: hh, fd_seek: Hv, fd_write: qv, setTempRet0: fh }, eT = dh(), jv = a.___wasm_call_ctors = function() {
        return (jv = a.___wasm_call_ctors = a.asm.__wasm_call_ctors).apply(null, arguments);
      }, mh = a._init = function() {
        return (mh = a._init = a.asm.init).apply(null, arguments);
      }, gh = a._init_with_threads_count = function() {
        return (gh = a._init_with_threads_count = a.asm.init_with_threads_count).apply(null, arguments);
      }, kc = a._get_threads_count = function() {
        return (kc = a._get_threads_count = a.asm.get_threads_count).apply(null, arguments);
      }, Sc = a._register_tensor = function() {
        return (Sc = a._register_tensor = a.asm.register_tensor).apply(null, arguments);
      }, bh = a._dispose_data = function() {
        return (bh = a._dispose_data = a.asm.dispose_data).apply(null, arguments);
      }, $e = a._dispose = function() {
        return ($e = a._dispose = a.asm.dispose).apply(null, arguments);
      }, yh = a._Abs = function() {
        return (yh = a._Abs = a.asm.Abs).apply(null, arguments);
      }, Ic = a._Add = function() {
        return (Ic = a._Add = a.asm.Add).apply(null, arguments);
      }, Mr = a._AddN = function() {
        return (Mr = a._AddN = a.asm.AddN).apply(null, arguments);
      }, Ei = a._All = function() {
        return (Ei = a._All = a.asm.All).apply(null, arguments);
      }, vh = a._Any = function() {
        return (vh = a._Any = a.asm.Any).apply(null, arguments);
      }, Kv = a._ArgMax = function() {
        return (Kv = a._ArgMax = a.asm.ArgMax).apply(null, arguments);
      }, xh = a._AvgPool = function() {
        return (xh = a._AvgPool = a.asm.AvgPool).apply(null, arguments);
      }, Xv = a._BatchMatMul = function() {
        return (Xv = a._BatchMatMul = a.asm.BatchMatMul).apply(null, arguments);
      }, Lr = a._Ceil = function() {
        return (Lr = a._Ceil = a.asm.Ceil).apply(null, arguments);
      }, wh = a._ClipByValue = function() {
        return (wh = a._ClipByValue = a.asm.ClipByValue).apply(null, arguments);
      }, kh = a._Conv2D = function() {
        return (kh = a._Conv2D = a.asm.Conv2D).apply(null, arguments);
      }, Sh = a._Conv2DBackpropInput = function() {
        return (Sh = a._Conv2DBackpropInput = a.asm.Conv2DBackpropInput).apply(null, arguments);
      }, Ih = a._Cos = function() {
        return (Ih = a._Cos = a.asm.Cos).apply(null, arguments);
      }, Ch = a._Cosh = function() {
        return (Ch = a._Cosh = a.asm.Cosh).apply(null, arguments);
      }, Nh = a._CropAndResize = function() {
        return (Nh = a._CropAndResize = a.asm.CropAndResize).apply(null, arguments);
      }, Cc = a._Cumprod = function() {
        return (Cc = a._Cumprod = a.asm.Cumprod).apply(null, arguments);
      }, Th = a._Cumsum = function() {
        return (Th = a._Cumsum = a.asm.Cumsum).apply(null, arguments);
      }, $h = a._DepthToSpace = function() {
        return ($h = a._DepthToSpace = a.asm.DepthToSpace).apply(null, arguments);
      }, _h = a._DepthwiseConv2dNative = function() {
        return (_h = a._DepthwiseConv2dNative = a.asm.DepthwiseConv2dNative).apply(null, arguments);
      }, Ah = a._Elu = function() {
        return (Ah = a._Elu = a.asm.Elu).apply(null, arguments);
      }, Eh = a._Equal = function() {
        return (Eh = a._Equal = a.asm.Equal).apply(null, arguments);
      }, Nc = a._Exp = function() {
        return (Nc = a._Exp = a.asm.Exp).apply(null, arguments);
      }, Rh = a._FlipLeftRight = function() {
        return (Rh = a._FlipLeftRight = a.asm.FlipLeftRight).apply(null, arguments);
      }, Dh = a._Floor = function() {
        return (Dh = a._Floor = a.asm.Floor).apply(null, arguments);
      }, Br = a._FloorDiv = function() {
        return (Br = a._FloorDiv = a.asm.FloorDiv).apply(null, arguments);
      }, wu = a._FusedBatchNorm = function() {
        return (wu = a._FusedBatchNorm = a.asm.FusedBatchNorm).apply(null, arguments);
      }, Fh = a._FusedConv2D = function() {
        return (Fh = a._FusedConv2D = a.asm.FusedConv2D).apply(null, arguments);
      }, Oh = a._FusedDepthwiseConv2D = function() {
        return (Oh = a._FusedDepthwiseConv2D = a.asm.FusedDepthwiseConv2D).apply(null, arguments);
      }, Ph = a._Gather = function() {
        return (Ph = a._Gather = a.asm.Gather).apply(null, arguments);
      }, Me = a._GatherNd = function() {
        return (Me = a._GatherNd = a.asm.GatherNd).apply(null, arguments);
      }, zh = a._Greater = function() {
        return (zh = a._Greater = a.asm.Greater).apply(null, arguments);
      }, Mh = a._GreaterEqual = function() {
        return (Mh = a._GreaterEqual = a.asm.GreaterEqual).apply(null, arguments);
      }, Lh = a._LeakyRelu = function() {
        return (Lh = a._LeakyRelu = a.asm.LeakyRelu).apply(null, arguments);
      }, Bh = a._Less = function() {
        return (Bh = a._Less = a.asm.Less).apply(null, arguments);
      }, Vh = a._LessEqual = function() {
        return (Vh = a._LessEqual = a.asm.LessEqual).apply(null, arguments);
      }, Wh = a._Log = function() {
        return (Wh = a._Log = a.asm.Log).apply(null, arguments);
      }, ku = a._LogicalAnd = function() {
        return (ku = a._LogicalAnd = a.asm.LogicalAnd).apply(null, arguments);
      }, Tc = a._Max = function() {
        return (Tc = a._Max = a.asm.Max).apply(null, arguments);
      }, $c = a._MaxPool = function() {
        return ($c = a._MaxPool = a.asm.MaxPool).apply(null, arguments);
      }, Uh = a._Maximum = function() {
        return (Uh = a._Maximum = a.asm.Maximum).apply(null, arguments);
      }, Gh = a._Mean = function() {
        return (Gh = a._Mean = a.asm.Mean).apply(null, arguments);
      }, Hh = a._Min = function() {
        return (Hh = a._Min = a.asm.Min).apply(null, arguments);
      }, qh = a._Minimum = function() {
        return (qh = a._Minimum = a.asm.Minimum).apply(null, arguments);
      }, jh = a._MirrorPad = function() {
        return (jh = a._MirrorPad = a.asm.MirrorPad).apply(null, arguments);
      }, Kh = a._Multiply = function() {
        return (Kh = a._Multiply = a.asm.Multiply).apply(null, arguments);
      }, gt = a._Neg = function() {
        return (gt = a._Neg = a.asm.Neg).apply(null, arguments);
      }, Xh = a._NonMaxSuppressionV3 = function() {
        return (Xh = a._NonMaxSuppressionV3 = a.asm.NonMaxSuppressionV3).apply(null, arguments);
      }, Yh = a._NonMaxSuppressionV4 = function() {
        return (Yh = a._NonMaxSuppressionV4 = a.asm.NonMaxSuppressionV4).apply(null, arguments);
      }, Qh = a._NonMaxSuppressionV5 = function() {
        return (Qh = a._NonMaxSuppressionV5 = a.asm.NonMaxSuppressionV5).apply(null, arguments);
      }, Ri = a._NotEqual = function() {
        return (Ri = a._NotEqual = a.asm.NotEqual).apply(null, arguments);
      }, _c = a._OneHot = function() {
        return (_c = a._OneHot = a.asm.OneHot).apply(null, arguments);
      }, Ac = a._PadV2 = function() {
        return (Ac = a._PadV2 = a.asm.PadV2).apply(null, arguments);
      }, Ec = a._Pow = function() {
        return (Ec = a._Pow = a.asm.Pow).apply(null, arguments);
      }, Zh = a._Prelu = function() {
        return (Zh = a._Prelu = a.asm.Prelu).apply(null, arguments);
      }, Rc = a._Prod = function() {
        return (Rc = a._Prod = a.asm.Prod).apply(null, arguments);
      }, Jh = a._RealDiv = function() {
        return (Jh = a._RealDiv = a.asm.RealDiv).apply(null, arguments);
      }, Yv = a._Relu = function() {
        return (Yv = a._Relu = a.asm.Relu).apply(null, arguments);
      }, Dc = a._Relu6 = function() {
        return (Dc = a._Relu6 = a.asm.Relu6).apply(null, arguments);
      }, Qv = a._ResizeBilinear = function() {
        return (Qv = a._ResizeBilinear = a.asm.ResizeBilinear).apply(null, arguments);
      }, ef = a._Reverse = function() {
        return (ef = a._Reverse = a.asm.Reverse).apply(null, arguments);
      }, tf = a._RotateWithOffset = function() {
        return (tf = a._RotateWithOffset = a.asm.RotateWithOffset).apply(null, arguments);
      }, nf = a._Round = function() {
        return (nf = a._Round = a.asm.Round).apply(null, arguments);
      }, sf = a._Rsqrt = function() {
        return (sf = a._Rsqrt = a.asm.Rsqrt).apply(null, arguments);
      }, rf = a._ScatterNd = function() {
        return (rf = a._ScatterNd = a.asm.ScatterNd).apply(null, arguments);
      }, af = a._SelectV2 = function() {
        return (af = a._SelectV2 = a.asm.SelectV2).apply(null, arguments);
      }, of = a._Sigmoid = function() {
        return (of = a._Sigmoid = a.asm.Sigmoid).apply(null, arguments);
      }, uf = a._Sin = function() {
        return (uf = a._Sin = a.asm.Sin).apply(null, arguments);
      }, lf = a._Softmax = function() {
        return (lf = a._Softmax = a.asm.Softmax).apply(null, arguments);
      }, cf = a._SparseFillEmptyRows = function() {
        return (cf = a._SparseFillEmptyRows = a.asm.SparseFillEmptyRows).apply(null, arguments);
      }, df = a._SparseReshape = function() {
        return (df = a._SparseReshape = a.asm.SparseReshape).apply(null, arguments);
      }, pf = a._SparseSegmentReduction = function() {
        return (pf = a._SparseSegmentReduction = a.asm.SparseSegmentReduction).apply(null, arguments);
      }, hf = a._Sqrt = function() {
        return (hf = a._Sqrt = a.asm.Sqrt).apply(null, arguments);
      }, ff = a._Square = function() {
        return (ff = a._Square = a.asm.Square).apply(null, arguments);
      }, mf = a._SquaredDifference = function() {
        return (mf = a._SquaredDifference = a.asm.SquaredDifference).apply(null, arguments);
      }, gf = a._Step = function() {
        return (gf = a._Step = a.asm.Step).apply(null, arguments);
      }, bf = a._StridedSlice = function() {
        return (bf = a._StridedSlice = a.asm.StridedSlice).apply(null, arguments);
      }, yf = a._Sub = function() {
        return (yf = a._Sub = a.asm.Sub).apply(null, arguments);
      }, vf = a._Sum = function() {
        return (vf = a._Sum = a.asm.Sum).apply(null, arguments);
      }, xf = a._Tan = function() {
        return (xf = a._Tan = a.asm.Tan).apply(null, arguments);
      }, wf = a._Tanh = function() {
        return (wf = a._Tanh = a.asm.Tanh).apply(null, arguments);
      }, kf = a._Tile = function() {
        return (kf = a._Tile = a.asm.Tile).apply(null, arguments);
      }, Sf = a._TopK = function() {
        return (Sf = a._TopK = a.asm.TopK).apply(null, arguments);
      }, If = a._Transform = function() {
        return (If = a._Transform = a.asm.Transform).apply(null, arguments);
      }, Cf = a._Transpose = function() {
        return (Cf = a._Transpose = a.asm.Transpose).apply(null, arguments);
      }, Nf = a.__FusedMatMul = function() {
        return (Nf = a.__FusedMatMul = a.asm._FusedMatMul).apply(null, arguments);
      }, Tf = a._malloc = function() {
        return (Tf = a._malloc = a.asm.malloc).apply(null, arguments);
      }, $f = a._free = function() {
        return ($f = a._free = a.asm.free).apply(null, arguments);
      }, _f = a.___errno_location = function() {
        return (_f = a.___errno_location = a.asm.__errno_location).apply(null, arguments);
      }, Af = a._emscripten_main_thread_process_queued_calls = function() {
        return (Af = a._emscripten_main_thread_process_queued_calls = a.asm.emscripten_main_thread_process_queued_calls).apply(null, arguments);
      }, Fc = a.stackSave = function() {
        return (Fc = a.stackSave = a.asm.stackSave).apply(null, arguments);
      }, Oc = a.stackRestore = function() {
        return (Oc = a.stackRestore = a.asm.stackRestore).apply(null, arguments);
      }, Su = a.stackAlloc = function() {
        return (Su = a.stackAlloc = a.asm.stackAlloc).apply(null, arguments);
      }, Ef = a.dynCall_iijjiiii = function() {
        return (Ef = a.dynCall_iijjiiii = a.asm.dynCall_iijjiiii).apply(null, arguments);
      }, Rf = a.dynCall_jiji = function() {
        return (Rf = a.dynCall_jiji = a.asm.dynCall_jiji).apply(null, arguments);
      };
      a.cwrap = Ie;
      var Di;
      function Iu(H) {
        this.name = "ExitStatus", this.message = "Program terminated with exit(" + H + ")", this.status = H;
      }
      Js = function H() {
        Di || Cu(), Di || (Js = H);
      };
      function Cu(H) {
        if (H = H || c, Fn > 0 || (cc(), Fn > 0))
          return;
        function ee() {
          Di || (Di = true, a.calledRun = true, !oe && (dc(), i(a), a.onRuntimeInitialized && a.onRuntimeInitialized(), pc()));
        }
        a.setStatus ? (a.setStatus("Running..."), setTimeout(function() {
          setTimeout(function() {
            a.setStatus("");
          }, 1), ee();
        }, 1)) : ee();
      }
      a.run = Cu;
      function Zv(H) {
        ae = H, bu() || (a.onExit && a.onExit(H), oe = true), d(H, new Iu(H));
      }
      if (a.preInit)
        for (typeof a.preInit == "function" && (a.preInit = [a.preInit]); a.preInit.length > 0; )
          a.preInit.pop()();
      Cu();
      var Fi;
      u && (Fi = { uncaughtException: process.listeners("uncaughtException").filter(function(H) {
        return !u.uncaughtException.indexOf(H) > -1;
      }), unhandledRejection: process.listeners("unhandledRejection").filter(function(H) {
        return !u.unhandledRejection.indexOf(H) > -1;
      }) });
      var Oi;
      if (typeof r != "undefined")
        Oi = r;
      else if (typeof WasmBackendModuleThreadedSimd != "undefined")
        Oi = WasmBackendModuleThreadedSimd;
      else
        throw new Error("Could not find wasm module in post.js");
      if (Fi) {
        var Df = Oi._dispose;
        Oi._dispose = function() {
          Df(), Fi.uncaughtException.forEach(function(H) {
            process.removeListener("uncaughtException", H);
          }), Fi.unhandledRejection.forEach(function(H) {
            process.removeListener("unhandledRejection", H);
          });
        };
      }
      return r.ready;
    };
  })();
  typeof e == "object" && typeof t == "object" ? t.exports = n : typeof define == "function" && define.amd ? define([], function() {
    return n;
  }) : typeof e == "object" && (e.WasmBackendModule = n);
} });
var m$ = 1e-7;
var g$ = 1e-4;
var Yd = class {
  constructor(e, t) {
    this.backend = e, this.dataMover = t, this.data = /* @__PURE__ */ new WeakMap(), this.dataIdsCount = 0;
  }
  get(e) {
    return this.data.has(e) || this.dataMover.moveData(this.backend, e), this.data.get(e);
  }
  set(e, t) {
    this.dataIdsCount++, this.data.set(e, t);
  }
  has(e) {
    return this.data.has(e);
  }
  delete(e) {
    return this.dataIdsCount--, this.data.delete(e);
  }
  numDataIds() {
    return this.dataIdsCount;
  }
};
var il = class {
  refCount(e) {
    return On("refCount");
  }
  incRef(e) {
    return On("incRef");
  }
  timerAvailable() {
    return true;
  }
  time(e) {
    return On("time");
  }
  read(e) {
    return On("read");
  }
  readSync(e) {
    return On("readSync");
  }
  readToGPU(e, t) {
    return On("readToGPU");
  }
  numDataIds() {
    return On("numDataIds");
  }
  disposeData(e, t) {
    return On("disposeData");
  }
  write(e, t, n) {
    return On("write");
  }
  move(e, t, n, s, r) {
    return On("move");
  }
  memory() {
    return On("memory");
  }
  floatPrecision() {
    return On("floatPrecision");
  }
  epsilon() {
    return this.floatPrecision() === 32 ? m$ : g$;
  }
  dispose() {
    return On("dispose");
  }
};
function On(e) {
  throw new Error(`'${e}' not yet implemented or not found in the registry. This kernel may not be supported by the tfjs backend you have chosen`);
}
function Jw(e) {
  let t = e.length, n = 0;
  for (; t > 0; )
    n = Math.random() * t | 0, t--, gd(e, t, n);
}
function b$(e, t) {
  if (e.length !== t.length)
    throw new Error(`Array sizes must match to be shuffled together First array length was ${e.length}Second array length was ${t.length}`);
  let n = e.length, s = 0;
  for (; n > 0; )
    s = Math.random() * n | 0, n--, gd(e, n, s), gd(t, n, s);
}
function Hu(e, t, n) {
  return Math.max(e, Math.min(t, n));
}
function y$(e) {
  return e % 2 === 0 ? e : e + 1;
}
function gd(e, t, n) {
  let s = e[t];
  e[t] = e[n], e[n] = s;
}
function v$(e) {
  let t = 0;
  for (let n = 0; n < e.length; n++)
    t += e[n];
  return t;
}
function x$(e, t) {
  let n = Math.random();
  return t * n + (1 - n) * e;
}
function w$(e, t) {
  let n = 0;
  for (let s = 0; s < e.length; s++) {
    let r = Number(e[s]) - Number(t[s]);
    n += r * r;
  }
  return n;
}
function F(e, t) {
  if (!e)
    throw new Error(typeof t == "string" ? t : t());
}
function pn(e, t, n = "") {
  F(kr(e, t), () => n + ` Shapes ${e} and ${t} must match`);
}
function ka(e) {
  F(e != null, () => "The input to the tensor constructor must be a non-null value.");
}
function ra(e, t = [], n = false) {
  if (t == null && (t = []), Array.isArray(e) || Qt(e) && !n)
    for (let s = 0; s < e.length; ++s)
      ra(e[s], t, n);
  else
    t.push(e);
  return t;
}
function dt(e) {
  if (e.length === 0)
    return 1;
  let t = e[0];
  for (let n = 1; n < e.length; n++)
    t *= e[n];
  return t;
}
function k$(e) {
  return e.length === 0;
}
function kr(e, t) {
  if (e === t)
    return true;
  if (e == null || t == null || e.length !== t.length)
    return false;
  for (let n = 0; n < e.length; n++)
    if (e[n] !== t[n])
      return false;
  return true;
}
function eo(e) {
  return e % 1 === 0;
}
function S$(e) {
  if (Math.tanh != null)
    return Math.tanh(e);
  if (e === 1 / 0)
    return 1;
  if (e === -1 / 0)
    return -1;
  {
    let t = Math.exp(2 * e);
    return (t - 1) / (t + 1);
  }
}
function I$(e) {
  let t = Math.ceil(Math.sqrt(e));
  return [t, Math.ceil(e / t)];
}
function C$(e) {
  let t = new Uint32Array(e);
  for (let n = 0; n < e; ++n)
    t[n] = n;
  return Jw(t), t;
}
function Vu(e, t) {
  return t <= e.length ? e : e + " ".repeat(t - e.length);
}
function N$(e, t = (s) => 0, n) {
  return new Promise((s, r) => {
    let a = 0, i = () => {
      if (e()) {
        s();
        return;
      }
      a++;
      let o = t(a);
      if (n != null && a >= n) {
        r();
        return;
      }
      setTimeout(i, o);
    };
    i();
  });
}
function T$(e, t) {
  let n = 1, s = -1;
  for (let a = 0; a < e.length; ++a)
    if (e[a] >= 0)
      n *= e[a];
    else if (e[a] === -1) {
      if (s !== -1)
        throw Error(`Shapes can only have 1 implicit size. Found -1 at dim ${s} and dim ${a}`);
      s = a;
    } else if (e[a] < 0)
      throw Error(`Shapes can not be < 0. Found ${e[a]} at dim ${a}`);
  if (s === -1) {
    if (t > 0 && t !== n)
      throw Error(`Size(${t}) must match the product of shape ${e}`);
    return e;
  }
  if (n === 0)
    throw Error(`Cannot infer the missing size in [${e}] when there are 0 elements`);
  if (t % n !== 0)
    throw Error(`The implicit shape can't be a fractional number. Got ${t} / ${n}`);
  let r = e.slice();
  return r[s] = t / n, r;
}
function ts(e, t) {
  let n = t.length;
  return e = e == null ? t.map((s, r) => r) : [].concat(e), F(e.every((s) => s >= -n && s < n), () => `All values in axis param must be in range [-${n}, ${n}) but got axis ${e}`), F(e.every((s) => eo(s)), () => `All values in axis param must be integers but got axis ${e}`), e.map((s) => s < 0 ? n + s : s);
}
function ek(e, t) {
  let n = [], s = [], r = t != null && Array.isArray(t) && t.length === 0, a = t == null || r ? null : ts(t, e).sort(), i = 0;
  for (let o = 0; o < e.length; ++o) {
    if (a != null) {
      if (a[i] === o && e[o] !== 1)
        throw new Error(`Can't squeeze axis ${o} since its dim '${e[o]}' is not 1`);
      (a[i] == null || a[i] > o) && e[o] === 1 && (n.push(e[o]), s.push(o)), a[i] <= o && i++;
    }
    e[o] !== 1 && (n.push(e[o]), s.push(o));
  }
  return { newShape: n, keptDims: s };
}
function tk(e, t) {
  let n = null;
  if (e == null || e === "float32")
    n = new Float32Array(t);
  else if (e === "int32")
    n = new Int32Array(t);
  else if (e === "bool")
    n = new Uint8Array(t);
  else
    throw new Error(`Unknown data type ${e}`);
  return n;
}
function nk(e, t) {
  let n = null;
  if (e == null || e === "float32")
    n = new Float32Array(t);
  else if (e === "int32")
    n = new Int32Array(t);
  else if (e === "bool")
    n = new Uint8Array(t);
  else if (e === "string")
    n = new Array(t);
  else
    throw new Error(`Unknown data type ${e}`);
  return n;
}
function sk(e, t) {
  for (let n = 0; n < e.length; n++) {
    let s = e[n];
    if (isNaN(s) || !isFinite(s))
      throw Error(`A tensor of type ${t} being uploaded contains ${s}.`);
  }
}
function rk(e) {
  return e === "bool" || e === "complex64" || e === "float32" || e === "int32" || e === "string";
}
function $$(e, t) {
  return !(t === "complex64" || t === "float32" && e !== "complex64" || t === "int32" && e !== "float32" && e !== "complex64" || t === "bool" && e === "bool");
}
function Qt(e) {
  return e instanceof Float32Array || e instanceof Int32Array || e instanceof Uint8Array || e instanceof Uint8ClampedArray;
}
function tm(e) {
  if (e === "float32" || e === "int32")
    return 4;
  if (e === "complex64")
    return 8;
  if (e === "bool")
    return 1;
  throw new Error(`Unknown dtype ${e}`);
}
function ak(e) {
  if (e == null)
    return 0;
  let t = 0;
  return e.forEach((n) => t += n.length), t;
}
function ar(e) {
  return typeof e == "string" || e instanceof String;
}
function ik(e) {
  return typeof e == "boolean";
}
function ok(e) {
  return typeof e == "number";
}
function Qd(e) {
  return Array.isArray(e) ? Qd(e[0]) : e instanceof Float32Array ? "float32" : e instanceof Int32Array || e instanceof Uint8Array || e instanceof Uint8ClampedArray ? "int32" : ok(e) ? "float32" : ar(e) ? "string" : ik(e) ? "bool" : "float32";
}
function hr(e) {
  return !!(e && e.constructor && e.call && e.apply);
}
function bd(e, t) {
  for (let n = t; n < e; ++n)
    if (e % n === 0)
      return n;
  return e;
}
function co(e) {
  let t = e.length;
  if (t < 2)
    return [];
  let n = new Array(t - 1);
  n[t - 2] = e[t - 1];
  for (let s = t - 3; s >= 0; --s)
    n[s] = n[s + 1] * e[s + 1];
  return n;
}
function uk(e, t, n, s = false) {
  let r = new Array();
  if (t.length === 1) {
    let a = t[0] * (s ? 2 : 1);
    for (let i = 0; i < a; i++)
      r[i] = n[e + i];
  } else {
    let a = t[0], i = t.slice(1), o = i.reduce((u, l) => u * l) * (s ? 2 : 1);
    for (let u = 0; u < a; u++)
      r[u] = uk(e + u * o, i, n, s);
  }
  return r;
}
function Xi(e, t, n = false) {
  if (e.length === 0)
    return t[0];
  let s = e.reduce((r, a) => r * a) * (n ? 2 : 1);
  if (s === 0)
    return [];
  if (s !== t.length)
    throw new Error(`[${e}] does not match the input size ${t.length}${n ? " for a complex tensor" : ""}.`);
  return uk(0, e, t, n);
}
function ug(e, t) {
  let n = Zd(e, t);
  for (let s = 0; s < n.length; s++)
    n[s] = 1;
  return n;
}
function Zd(e, t) {
  if (t == null || t === "float32" || t === "complex64")
    return new Float32Array(e);
  if (t === "int32")
    return new Int32Array(e);
  if (t === "bool")
    return new Uint8Array(e);
  throw new Error(`Unknown data type ${t}`);
}
function _$(e, t) {
  let n = e.reduce((s, r) => s * r, 1);
  if (t == null || t === "float32")
    return Xi(e, new Float32Array(n));
  if (t === "int32")
    return Xi(e, new Int32Array(n));
  if (t === "bool")
    return Xi(e, new Uint8Array(n));
  throw new Error(`Unknown data type ${t}`);
}
function lg(e) {
  e.forEach((t) => {
    F(Number.isInteger(t) && t >= 0, () => `Tensor must have a shape comprised of positive integers but got shape [${e}].`);
  });
}
function A$(e, t, n) {
  if (t === 0)
    return 0;
  if (t === 1)
    return e[0];
  let s = e[e.length - 1];
  for (let r = 0; r < e.length - 1; ++r)
    s += n[r] * e[r];
  return s;
}
function E$(e, t, n) {
  if (t === 0)
    return [];
  if (t === 1)
    return [e];
  let s = new Array(t);
  for (let r = 0; r < s.length - 1; ++r)
    s[r] = Math.floor(e / n[r]), e -= s[r] * n[r];
  return s[s.length - 1] = e, s;
}
function cg(e) {
  return e && e.then && typeof e.then == "function";
}
var ux = "tfjsflags";
var R$ = class {
  constructor(e) {
    this.global = e, this.flags = {}, this.flagRegistry = {}, this.urlFlags = {}, this.getQueryParams = D$, this.populateURLFlags();
  }
  setPlatform(e, t) {
    this.platform != null && (K().getBool("IS_TEST") || K().getBool("PROD") || console.warn(`Platform ${this.platformName} has already been set. Overwriting the platform with ${e}.`)), this.platformName = e, this.platform = t;
  }
  registerFlag(e, t, n) {
    if (this.flagRegistry[e] = { evaluationFn: t, setHook: n }, this.urlFlags[e] != null) {
      let s = this.urlFlags[e];
      K().getBool("IS_TEST") || K().getBool("PROD") || console.warn(`Setting feature override from URL ${e}: ${s}.`), this.set(e, s);
    }
  }
  async getAsync(e) {
    return e in this.flags ? this.flags[e] : (this.flags[e] = await this.evaluateFlag(e), this.flags[e]);
  }
  get(e) {
    if (e in this.flags)
      return this.flags[e];
    let t = this.evaluateFlag(e);
    if (cg(t))
      throw new Error(`Flag ${e} cannot be synchronously evaluated. Please use getAsync() instead.`);
    return this.flags[e] = t, this.flags[e];
  }
  getNumber(e) {
    return this.get(e);
  }
  getBool(e) {
    return this.get(e);
  }
  getFlags() {
    return this.flags;
  }
  get features() {
    return this.flags;
  }
  set(e, t) {
    if (this.flagRegistry[e] == null)
      throw new Error(`Cannot set flag ${e} as it has not been registered.`);
    this.flags[e] = t, this.flagRegistry[e].setHook != null && this.flagRegistry[e].setHook(t);
  }
  evaluateFlag(e) {
    if (this.flagRegistry[e] == null)
      throw new Error(`Cannot evaluate flag '${e}': no evaluation function found.`);
    return this.flagRegistry[e].evaluationFn();
  }
  setFlags(e) {
    this.flags = Object.assign({}, e);
  }
  reset() {
    this.flags = {}, this.urlFlags = {}, this.populateURLFlags();
  }
  populateURLFlags() {
    if (typeof this.global == "undefined" || typeof this.global.location == "undefined" || typeof this.global.location.search == "undefined")
      return;
    let e = this.getQueryParams(this.global.location.search);
    ux in e && e[ux].split(",").forEach((n) => {
      let [s, r] = n.split(":");
      this.urlFlags[s] = O$(s, r);
    });
  }
};
function D$(e) {
  let t = {};
  return e.replace(/[?&]([^=?&]+)(?:=([^&]*))?/g, (n, ...s) => (F$(t, s[0], s[1]), s.join("="))), t;
}
function F$(e, t, n) {
  e[decodeURIComponent(t)] = decodeURIComponent(n || "");
}
function O$(e, t) {
  if (t = t.toLowerCase(), t === "true" || t === "false")
    return t === "true";
  if (`${+t}` === t)
    return +t;
  throw new Error(`Could not parse value flag value ${t} for flag ${e}.`);
}
function K() {
  return lk;
}
var lk = null;
function P$(e) {
  lk = e;
}
var Bf;
function ck() {
  if (Bf == null) {
    let e;
    if (typeof window != "undefined")
      e = window;
    else if (typeof global != "undefined")
      e = global;
    else if (typeof process != "undefined")
      e = process;
    else if (typeof self != "undefined")
      e = self;
    else
      throw new Error("Could not find a global object");
    Bf = e;
  }
  return Bf;
}
function z$() {
  let e = ck();
  return e._tfGlobals == null && (e._tfGlobals = /* @__PURE__ */ new Map()), e._tfGlobals;
}
function dg(e, t) {
  let n = z$();
  if (n.has(e))
    return n.get(e);
  {
    let s = t();
    return n.set(e, s), n.get(e);
  }
}
var po = "Abs";
var ol = "Acos";
var ul = "Acosh";
var Sr = "Add";
var Sa = "AddN";
var ll = "All";
var cl = "Any";
var Ia = "ArgMax";
var dl = "ArgMin";
var pl = "Asin";
var hl = "Asinh";
var fl = "Atan";
var ml = "Atanh";
var gl = "Atan2";
var Ca = "AvgPool";
var pg = "AvgPoolGrad";
var Jd = "AvgPool3D";
var hg = "AvgPool3DGrad";
var Na = "BatchMatMul";
var ho = "BatchToSpaceND";
var fg = "Bincount";
var M$ = "BroadcastTo";
var mg = "BroadcastArgs";
var Ta = "Cast";
var $a = "Ceil";
var Ir = "ClipByValue";
var ep = "Complex";
var tp = "ComplexAbs";
var fo = "Concat";
var _a = "Conv2D";
var gg = "Conv2DBackpropFilter";
var Aa = "Conv2DBackpropInput";
var np = "Conv3D";
var bg = "Conv3DBackpropFilterV2";
var yg = "Conv3DBackpropInputV2";
var Ea = "Cos";
var Ra = "Cosh";
var mo = "Cumprod";
var Da = "Cumsum";
var go = "CropAndResize";
var vg = "DenseBincount";
var bo = "DepthToSpace";
var Fa = "DepthwiseConv2dNative";
var xg = "DepthwiseConv2dNativeBackpropFilter";
var wg = "DepthwiseConv2dNativeBackpropInput";
var kg = "Diag";
var sp = "Dilation2D";
var nm = "Dilation2DBackpropInput";
var sm = "Dilation2DBackpropFilter";
var Oa = "RealDiv";
var rp = "Einsum";
var Pa = "Elu";
var Sg = "EluGrad";
var bl = "Erf";
var yo = "Equal";
var za = "Exp";
var vo = "ExpandDims";
var xo = "Expm1";
var Ig = "FFT";
var yl = "Fill";
var wo = "FlipLeftRight";
var Ma = "Floor";
var La = "FloorDiv";
var Ba = "FusedBatchNorm";
var ko = "GatherV2";
var So = "GatherNd";
var Io = "Greater";
var Va = "GreaterEqual";
var Wa = "Identity";
var Cg = "IFFT";
var ap = "Imag";
var vl = "IsFinite";
var xl = "IsInf";
var wl = "IsNan";
var Ua = "LeakyRelu";
var Co = "Less";
var No = "LessEqual";
var Ng = "LinSpace";
var Ga = "Log";
var kl = "Log1p";
var To = "LogicalAnd";
var Sl = "LogicalNot";
var ip = "LogicalOr";
var L$ = "LogSoftmax";
var rpe = "LowerBound";
var op = "LRN";
var Tg = "LRNGrad";
var Ha = "Max";
var qa = "Maximum";
var ja = "MaxPool";
var $g = "MaxPoolGrad";
var up = "MaxPool3D";
var _g = "MaxPool3DGrad";
var Ag = "MaxPoolWithArgmax";
var Ka = "Mean";
var Xa = "Min";
var Ya = "Minimum";
var Qa = "MirrorPad";
var Il = "Mod";
var Eg = "Multinomial";
var Za = "Multiply";
var $o = "Neg";
var _o = "NotEqual";
var Ao = "NonMaxSuppressionV3";
var Cl = "NonMaxSuppressionV4";
var Eo = "NonMaxSuppressionV5";
var Ro = "OnesLike";
var Do = "OneHot";
var Fo = "Pack";
var Ja = "PadV2";
var ape = "Pool";
var ei = "Pow";
var ti = "Prelu";
var ni = "Prod";
var Nl = "Range";
var lp = "Real";
var Tl = "Reciprocal";
var si = "Relu";
var Oo = "Reshape";
var $l = "ResizeNearestNeighbor";
var Rg = "ResizeNearestNeighborGrad";
var ri = "ResizeBilinear";
var Dg = "ResizeBilinearGrad";
var ai = "Relu6";
var Po = "Reverse";
var zo = "Round";
var ii = "Rsqrt";
var Mo = "ScatterNd";
var Fg = "SearchSorted";
var Lo = "Select";
var _l = "Selu";
var Bo = "Slice";
var oi = "Sin";
var Vo = "Sinh";
var Al = "Sign";
var ui = "Sigmoid";
var El = "Softplus";
var li = "Sqrt";
var ci = "Sum";
var Wo = "SpaceToBatchND";
var Uo = "SplitV";
var di = "Softmax";
var cp = "SparseFillEmptyRows";
var Rl = "SparseReshape";
var dp = "SparseSegmentMean";
var pp = "SparseSegmentSum";
var hp = "SparseToDense";
var pi = "SquaredDifference";
var Dl = "Square";
var Go = "StridedSlice";
var fp = "StringNGrams";
var Og = "StringSplit";
var Pg = "StringToHashBucketFast";
var hi = "Sub";
var Ho = "Tan";
var fi = "Tanh";
var Cr = "Tile";
var qo = "TopK";
var jo = "Transform";
var mi = "Transpose";
var zg = "Unique";
var Ko = "Unpack";
var mp = "UnsortedSegmentSum";
var ipe = "UpperBound";
var Xo = "ZerosLike";
var gi = "Step";
var yd = "FromPixels";
var Yo = "RotateWithOffset";
var aa = "_FusedMatMul";
var ia = "FusedConv2D";
var oa = "FusedDepthwiseConv2D";
function rr(...e) {
  K().getBool("IS_TEST") || K().getBool("PROD") || console.warn(...e);
}
function B$(...e) {
  K().getBool("IS_TEST") || K().getBool("PROD") || console.log(...e);
}
var to = dg("kernelRegistry", () => /* @__PURE__ */ new Map());
var qu = dg("gradRegistry", () => /* @__PURE__ */ new Map());
function rm(e, t) {
  let n = Mg(e, t);
  return to.get(n);
}
function lx(e) {
  return qu.get(e);
}
function am(e) {
  let t = to.entries(), n = [];
  for (; ; ) {
    let { done: s, value: r } = t.next();
    if (s)
      break;
    let [a, i] = r, [o] = a.split("_");
    o === e && n.push(i);
  }
  return n;
}
function Fl(e) {
  let { kernelName: t, backendName: n } = e, s = Mg(t, n);
  to.has(s) && rr(`The kernel '${t}' for backend '${n}' is already registered`), to.set(s, e);
}
function V$(e) {
  let { kernelName: t } = e;
  qu.has(t) && K().getBool("DEBUG") && rr(`Overriding the gradient for '${t}'`), qu.set(t, e);
}
function ope(e, t) {
  let n = Mg(e, t);
  if (!to.has(n))
    throw new Error(`The kernel '${e}' for backend '${t}' is not registered`);
  to.delete(n);
}
function upe(e) {
  if (!qu.has(e))
    throw new Error(`The gradient '${e}' for backend is not registered`);
  qu.delete(e);
}
function lpe(e, t) {
  am(e).forEach((s) => {
    let r = Object.assign({}, s, { backendName: t });
    Fl(r);
  });
}
function Mg(e, t) {
  return `${t}_${e}`;
}
var w = {};
Ee(w, { arraysEqual: () => kr, assert: () => F, assertNonNegativeIntegerDimensions: () => lg, assertNonNull: () => ka, assertShapesMatch: () => pn, bytesFromStringArray: () => ak, bytesPerElement: () => tm, checkConversionForErrors: () => sk, clamp: () => Hu, computeStrides: () => co, createScalarValue: () => j$, createShuffledIndices: () => C$, decodeString: () => vd, distSquared: () => w$, encodeString: () => Pl, fetch: () => X$, fingerPrint64: () => q$, flatten: () => ra, getArrayFromDType: () => nk, getTypedArrayFromDType: () => tk, hasEncodingLoss: () => $$, hexToLong: () => Ol, indexToLoc: () => E$, inferDtype: () => Qd, inferFromImplicitShape: () => T$, isBoolean: () => ik, isFunction: () => hr, isInt: () => eo, isNumber: () => ok, isPromise: () => cg, isScalarShape: () => k$, isString: () => ar, isTypedArray: () => Qt, isValidDtype: () => rk, locToIndex: () => A$, makeOnesTypedArray: () => ug, makeZerosNestedTypedArray: () => _$, makeZerosTypedArray: () => Zd, nearestDivisor: () => bd, nearestLargerEven: () => y$, now: () => ju, parseAxisParam: () => ts, randUniform: () => x$, repeatedTry: () => N$, rightPad: () => Vu, shuffle: () => Jw, shuffleCombo: () => b$, sizeFromShape: () => dt, sizeToSquarishShape: () => I$, squeezeShape: () => ek, sum: () => v$, swap: () => gd, tanh: () => S$, toNestedArray: () => Xi, toTypedArray: () => gp });
var cx = wa(JT());
var jr = cx.default || cx;
function Ol(e) {
  return jr.fromString(e, true, 16);
}
var dk = Ol("c3a5c85c97cb3127");
var Hr = Ol("b492b66fbe98f273");
var on = Ol("9ae16a3b2f90404f");
function im(e) {
  return e.xor(e.shru(47));
}
function pk(e, t, n) {
  let s = e.slice(t, t + n);
  return jr.fromBytes(Array.from(s), true, true);
}
function lt(e, t) {
  return pk(e, t, 8);
}
function dx(e, t) {
  return pk(e, t, 4);
}
function Bt(e, t) {
  return t === 0 ? e : e.shru(t).or(e.shl(64 - t));
}
function or(e, t, n = Ol("9ddfea08eb382d69")) {
  let s = e.xor(t).mul(n);
  s = s.xor(s.shru(47));
  let r = t.xor(s).mul(n);
  return r = r.xor(r.shru(47)), r = r.mul(n), r;
}
function W$(e, t, n, s, r, a) {
  r = r.add(e), a = Bt(a.add(r).add(s), 21);
  let i = r;
  return r = r.add(t), r = r.add(n), a = a.add(Bt(r, 44)), [r.add(s), a.add(i)];
}
function Uc(e, t, n, s) {
  return W$(lt(e, t), lt(e, t + 8), lt(e, t + 16), lt(e, t + 24), n, s);
}
function U$(e, t = e.length) {
  if (t >= 8) {
    let n = on.add(t * 2), s = lt(e, 0).add(on), r = lt(e, t - 8), a = Bt(r, 37).mul(n).add(s), i = Bt(s, 25).add(r).mul(n);
    return or(a, i, n);
  }
  if (t >= 4) {
    let n = on.add(t * 2), s = dx(e, 0);
    return or(s.shl(3).add(t), dx(e, t - 4), n);
  }
  if (t > 0) {
    let n = e[0], s = e[t >> 1], r = e[t - 1], a = n + (s << 8), i = t + (r << 2);
    return im(on.mul(a).xor(dk.mul(i))).mul(on);
  }
  return on;
}
function G$(e, t = e.length) {
  let n = on.add(t * 2), s = lt(e, 0).mul(Hr), r = lt(e, 8), a = lt(e, t - 8).mul(n), i = lt(e, t - 16).mul(on);
  return or(Bt(s.add(r), 43).add(Bt(a, 30)).add(i), s.add(Bt(r.add(on), 18)).add(a), n);
}
function H$(e, t = e.length) {
  let n = on.add(t * 2), s = lt(e, 0).mul(on), r = lt(e, 8), a = lt(e, t - 8).mul(n), i = lt(e, t - 16).mul(on), o = Bt(s.add(r), 43).add(Bt(a, 30)).add(i), u = or(o, s.add(Bt(r.add(on), 18)).add(a), n), l = lt(e, 16).mul(n), c = lt(e, 24), p = o.add(lt(e, t - 32)).mul(n), d = u.add(lt(e, t - 24)).mul(n);
  return or(Bt(l.add(c), 43).add(Bt(p, 30)).add(d), l.add(Bt(c.add(s), 18)).add(p), n);
}
function q$(e, t = e.length) {
  let n = jr.fromNumber(81, true);
  if (t <= 32)
    return t <= 16 ? U$(e, t) : G$(e, t);
  if (t <= 64)
    return H$(e, t);
  let s = n, r = n.mul(Hr).add(113), a = im(r.mul(on).add(113)).mul(on), i = [jr.UZERO, jr.UZERO], o = [jr.UZERO, jr.UZERO];
  s = s.mul(on).add(lt(e, 0));
  let u = 0, l = (t - 1 >> 6) * 64, c = l + (t - 1 & 63) - 63;
  do
    s = Bt(s.add(r).add(i[0]).add(lt(e, u + 8)), 37).mul(Hr), r = Bt(r.add(i[1]).add(lt(e, u + 48)), 42).mul(Hr), s = s.xor(o[1]), r = r.add(i[0]).add(lt(e, u + 40)), a = Bt(a.add(o[0]), 33).mul(Hr), i = Uc(e, u, i[1].mul(Hr), s.add(o[0])), o = Uc(e, u + 32, a.add(o[1]), r.add(lt(e, u + 16))), [a, s] = [s, a], u += 64;
  while (u !== l);
  let p = Hr.add(a.and(255).shl(1));
  return u = c, o[0] = o[0].add(t - 1 & 63), i[0] = i[0].add(o[0]), o[0] = o[0].add(i[0]), s = Bt(s.add(r).add(i[0]).add(lt(e, u + 8)), 37).mul(p), r = Bt(r.add(i[1]).add(lt(e, u + 48)), 42).mul(p), s = s.xor(o[1].mul(9)), r = r.add(i[0].mul(9).add(lt(e, u + 40))), a = Bt(a.add(o[0]), 33).mul(p), i = Uc(e, u, i[1].mul(p), s.add(o[0])), o = Uc(e, u + 32, a.add(o[1]), r.add(lt(e, u + 16))), [a, s] = [s, a], or(or(i[0], o[0], p).add(im(r).mul(dk)).add(a), or(i[1], o[1], p).add(s), p);
}
function j$(e, t) {
  return t === "string" ? Pl(e) : gp([e], t);
}
function K$(e, t) {
  return e instanceof Float32Array && t === "float32" || e instanceof Int32Array && t === "int32" || e instanceof Uint8Array && t === "bool";
}
function gp(e, t) {
  if (t === "string")
    throw new Error("Cannot convert a string[] to a TypedArray");
  if (Array.isArray(e) && (e = ra(e)), K().getBool("DEBUG") && sk(e, t), K$(e, t))
    return e;
  if (t == null || t === "float32" || t === "complex64")
    return new Float32Array(e);
  if (t === "int32")
    return new Int32Array(e);
  if (t === "bool") {
    let n = new Uint8Array(e.length);
    for (let s = 0; s < n.length; ++s)
      Math.round(e[s]) !== 0 && (n[s] = 1);
    return n;
  } else
    throw new Error(`Unknown data type ${t}`);
}
function ju() {
  return K().platform.now();
}
function X$(e, t) {
  return K().platform.fetch(e, t);
}
function Pl(e, t = "utf-8") {
  return t = t || "utf-8", K().platform.encode(e, t);
}
function vd(e, t = "utf-8") {
  return t = t || "utf-8", K().platform.decode(e, t);
}
var Y$ = class {
  constructor(e, t) {
    this.backendTimer = e, this.logger = t, t == null && (this.logger = new Z$());
  }
  profileKernel(e, t, n) {
    let s, r = () => {
      s = n();
    }, a, i = ju();
    if (this.backendTimer.timerAvailable())
      a = this.backendTimer.time(r);
    else {
      r();
      for (let u of s)
        u.dataSync();
      a = Promise.resolve({ kernelMs: ju() - i });
    }
    if (K().getBool("CHECK_COMPUTATION_FOR_ERRORS"))
      for (let u = 0; u < s.length; u++) {
        let l = s[u];
        l.data().then((c) => {
          Q$(c, l.dtype, e);
        });
      }
    return { kernelName: e, outputs: s, inputs: t, timeMs: a.then((u) => u.kernelMs), extraInfo: a.then((u) => u.getExtraProfileInfo != null ? u.getExtraProfileInfo() : "") };
  }
  logKernelProfile(e) {
    let { kernelName: t, outputs: n, timeMs: s, inputs: r, extraInfo: a } = e;
    n.forEach((i) => {
      Promise.all([i.data(), s, a]).then((o) => {
        this.logger.logKernelProfile(t, i, o[0], o[1], r, o[2]);
      });
    });
  }
};
function Q$(e, t, n) {
  if (t !== "float32")
    return false;
  for (let s = 0; s < e.length; s++) {
    let r = e[s];
    if (isNaN(r) || !isFinite(r))
      return console.warn(`Found ${r} in the result of '${n}'`), true;
  }
  return false;
}
var Z$ = class {
  logKernelProfile(e, t, n, s, r, a) {
    let i = typeof s == "number" ? Vu(`${s}ms`, 9) : s.error, o = Vu(e, 25), u = t.rank, l = t.size, c = Vu(t.shape.toString(), 14), p = "";
    for (let d in r) {
      let h = r[d];
      if (h != null) {
        let f = h.shape || t.shape, m = f.length;
        p += `${d}: ${m}D ${m > 0 ? f : ""} `;
      }
    }
    console.log(`%c${o}	%c${i}	%c${u}D ${c}	%c${l}	%c${p}	%c${a}`, "font-weight:bold", "color:red", "color:blue", "color: orange", "color: green", "color: steelblue");
  }
};
function J$(e, t, n) {
  let s = {}, r = {};
  for (let u = 0; u < t.length; u++)
    s[t[u].id] = true;
  for (let u = 0; u < e.length; u++) {
    let l = e[u], c = l.inputs;
    for (let p in c) {
      let d = c[p], h = false;
      for (let f = 0; f < t.length; f++)
        if (s[d.id]) {
          l.outputs.forEach((m) => s[m.id] = true), h = true, r[l.id] = true;
          break;
        }
      if (h)
        break;
    }
  }
  let a = {};
  a[n.id] = true;
  let i = {};
  for (let u = e.length - 1; u >= 0; u--) {
    let l = e[u], c = l.inputs;
    for (let p = 0; p < l.outputs.length; p++)
      if (a[l.outputs[p].id]) {
        for (let d in c)
          a[c[d].id] = true, i[l.id] = true;
        break;
      }
  }
  let o = [];
  for (let u = 0; u < e.length; u++) {
    let l = e[u];
    if (r[l.id] && i[l.id]) {
      let c = {};
      for (let d in l.inputs) {
        let h = l.inputs[d];
        s[h.id] && (c[d] = h);
      }
      let p = Object.assign({}, l);
      p.inputs = c, p.outputs = l.outputs, o.push(p);
    }
  }
  return o;
}
function e_(e, t, n, s) {
  for (let r = t.length - 1; r >= 0; r--) {
    let a = t[r], i = [];
    if (a.outputs.forEach((u) => {
      let l = e[u.id];
      l != null ? i.push(l) : i.push(null);
    }), a.gradient == null)
      throw new Error(`Cannot compute gradient: gradient function not found for ${a.kernelName}.`);
    let o = a.gradient(i);
    for (let u in a.inputs) {
      if (!(u in o))
        throw new Error(`Cannot backprop through input ${u}. Available gradients found: ${Object.keys(o)}.`);
      let l = n(() => o[u]());
      if (l.dtype !== "float32")
        throw new Error(`Error in gradient for op ${a.kernelName}. The gradient of input ${u} must have 'float32' dtype, but has '${l.dtype}'`);
      let c = a.inputs[u];
      if (!kr(l.shape, c.shape))
        throw new Error(`Error in gradient for op ${a.kernelName}. The gradient of input '${u}' has shape '${l.shape}', which does not match the shape of the input '${c.shape}'`);
      if (e[c.id] == null)
        e[c.id] = l;
      else {
        let p = e[c.id];
        e[c.id] = s(p, l), p.dispose();
      }
    }
  }
}
var px = 20;
var $u = 3;
var Vf = 7;
function t_(e, t, n, s) {
  let r = co(t), a = n_(e, t, n, r), i = t.length, o = sd(e, t, n, r, a), u = ["Tensor"];
  return s && (u.push(`  dtype: ${n}`), u.push(`  rank: ${i}`), u.push(`  shape: [${t}]`), u.push("  values:")), u.push(o.map((l) => "    " + l).join(`
`)), u.join(`
`);
}
function n_(e, t, n, s) {
  let r = dt(t), a = s[s.length - 1], i = new Array(a).fill(0), o = t.length, u = n === "complex64" ? Du(e) : e;
  if (o > 1)
    for (let l = 0; l < r / a; l++) {
      let c = l * a;
      for (let p = 0; p < a; p++)
        i[p] = Math.max(i[p], Ru(u[c + p], 0, n).length);
    }
  return i;
}
function Ru(e, t, n) {
  let s;
  return Array.isArray(e) ? s = `${parseFloat(e[0].toFixed(Vf))} + ${parseFloat(e[1].toFixed(Vf))}j` : ar(e) ? s = `'${e}'` : n === "bool" ? s = hk(e) : s = parseFloat(e.toFixed(Vf)).toString(), Vu(s, t);
}
function hk(e) {
  return e === 0 ? "false" : "true";
}
function sd(e, t, n, s, r, a = true) {
  let i = n === "complex64" ? 2 : 1, o = t[0], u = t.length;
  if (u === 0) {
    if (n === "complex64") {
      let m = Du(e);
      return [Ru(m[0], 0, n)];
    }
    return n === "bool" ? [hk(e[0])] : [e[0].toString()];
  }
  if (u === 1) {
    if (o > px) {
      let g = $u * i, b = Array.from(e.slice(0, g)), y = Array.from(e.slice((o - $u) * i, o * i));
      return n === "complex64" && (b = Du(b), y = Du(y)), ["[" + b.map((v, x) => Ru(v, r[x], n)).join(", ") + ", ..., " + y.map((v, x) => Ru(v, r[o - $u + x], n)).join(", ") + "]"];
    }
    let m = n === "complex64" ? Du(e) : Array.from(e);
    return ["[" + m.map((g, b) => Ru(g, r[b], n)).join(", ") + "]"];
  }
  let l = t.slice(1), c = s.slice(1), p = s[0] * i, d = [];
  if (o > px) {
    for (let m = 0; m < $u; m++) {
      let g = m * p, b = g + p;
      d.push(...sd(e.slice(g, b), l, n, c, r, false));
    }
    d.push("...");
    for (let m = o - $u; m < o; m++) {
      let g = m * p, b = g + p;
      d.push(...sd(e.slice(g, b), l, n, c, r, m === o - 1));
    }
  } else
    for (let m = 0; m < o; m++) {
      let g = m * p, b = g + p;
      d.push(...sd(e.slice(g, b), l, n, c, r, m === o - 1));
    }
  let h = u === 2 ? "," : "";
  d[0] = "[" + d[0] + h;
  for (let m = 1; m < d.length - 1; m++)
    d[m] = " " + d[m] + h;
  let f = `,
`;
  for (let m = 2; m < u; m++)
    f += `
`;
  return d[d.length - 1] = " " + d[d.length - 1] + "]" + (a ? "" : f), d;
}
function Du(e) {
  let t = [];
  for (let n = 0; n < e.length; n += 2)
    t.push([e[n], e[n + 1]]);
  return t;
}
var Wt = class {
  constructor(e, t, n) {
    if (this.dtype = t, this.shape = e.slice(), this.size = dt(e), n != null) {
      let s = n.length;
      F(s === this.size, () => `Length of values '${s}' does not match the size inferred by the shape '${this.size}'.`);
    }
    if (t === "complex64")
      throw new Error("complex64 dtype TensorBuffers are not supported. Please create a TensorBuffer for the real and imaginary parts separately and call tf.complex(real, imag).");
    this.values = n || nk(t, this.size), this.strides = co(e);
  }
  set(e, ...t) {
    t.length === 0 && (t = [0]), F(t.length === this.rank, () => `The number of provided coordinates (${t.length}) must match the rank (${this.rank})`);
    let n = this.locToIndex(t);
    this.values[n] = e;
  }
  get(...e) {
    e.length === 0 && (e = [0]);
    let t = 0;
    for (let s of e) {
      if (s < 0 || s >= this.shape[t]) {
        let r = `Requested out of range element at ${e}.   Buffer shape=${this.shape}`;
        throw new Error(r);
      }
      t++;
    }
    let n = e[e.length - 1];
    for (let s = 0; s < e.length - 1; ++s)
      n += this.strides[s] * e[s];
    return this.values[n];
  }
  locToIndex(e) {
    if (this.rank === 0)
      return 0;
    if (this.rank === 1)
      return e[0];
    let t = e[e.length - 1];
    for (let n = 0; n < e.length - 1; ++n)
      t += this.strides[n] * e[n];
    return t;
  }
  indexToLoc(e) {
    if (this.rank === 0)
      return [];
    if (this.rank === 1)
      return [e];
    let t = new Array(this.shape.length);
    for (let n = 0; n < t.length - 1; ++n)
      t[n] = Math.floor(e / this.strides[n]), e -= t[n] * this.strides[n];
    return t[t.length - 1] = e, t;
  }
  get rank() {
    return this.shape.length;
  }
  toTensor() {
    return cs().makeTensor(this.values, this.shape, this.dtype);
  }
};
var cs = null;
var Hi = null;
var s_ = null;
function r_(e) {
  cs = e;
}
function a_(e) {
  Hi = e;
}
function i_(e) {
  s_ = e;
}
var et = class {
  constructor(e, t, n, s) {
    this.kept = false, this.isDisposedInternal = false, this.shape = e.slice(), this.dtype = t || "float32", this.size = dt(e), this.strides = co(e), this.dataId = n, this.id = s, this.rankType = this.rank < 5 ? this.rank.toString() : "higher";
  }
  get rank() {
    return this.shape.length;
  }
  async buffer() {
    let e = await this.data();
    return Hi.buffer(this.shape, this.dtype, e);
  }
  bufferSync() {
    return Hi.buffer(this.shape, this.dtype, this.dataSync());
  }
  async array() {
    let e = await this.data();
    return Xi(this.shape, e, this.dtype === "complex64");
  }
  arraySync() {
    return Xi(this.shape, this.dataSync(), this.dtype === "complex64");
  }
  async data() {
    this.throwIfDisposed();
    let e = cs().read(this.dataId);
    if (this.dtype === "string") {
      let t = await e;
      try {
        return t.map((n) => vd(n));
      } catch (n) {
        throw new Error("Failed to decode the string bytes into utf-8. To get the original bytes, call tensor.bytes().");
      }
    }
    return e;
  }
  dataToGPU(e) {
    return this.throwIfDisposed(), cs().readToGPU(this.dataId, e);
  }
  dataSync() {
    this.throwIfDisposed();
    let e = cs().readSync(this.dataId);
    if (this.dtype === "string")
      try {
        return e.map((t) => vd(t));
      } catch (t) {
        throw new Error("Failed to decode the string bytes into utf-8. To get the original bytes, call tensor.bytes().");
      }
    return e;
  }
  async bytes() {
    this.throwIfDisposed();
    let e = await cs().read(this.dataId);
    return this.dtype === "string" ? e : new Uint8Array(e.buffer);
  }
  dispose() {
    this.isDisposed || (cs().disposeTensor(this), this.isDisposedInternal = true);
  }
  get isDisposed() {
    return this.isDisposedInternal;
  }
  throwIfDisposed() {
    if (this.isDisposed)
      throw new Error("Tensor is disposed.");
  }
  print(e = false) {
    return Hi.print(this, e);
  }
  clone() {
    return this.throwIfDisposed(), Hi.clone(this);
  }
  toString(e = false) {
    let t = this.dataSync();
    return t_(t, this.shape, this.dtype, e);
  }
  cast(e) {
    return this.throwIfDisposed(), Hi.cast(this, e);
  }
  variable(e = true, t, n) {
    return this.throwIfDisposed(), cs().makeVariable(this, e, t, n);
  }
};
Object.defineProperty(et, Symbol.hasInstance, { value: (e) => !!e && e.data != null && e.dataSync != null && e.throwIfDisposed != null });
function o_() {
  return dg("Tensor", () => et);
}
o_();
var xd = class extends et {
  constructor(e, t, n, s) {
    super(e.shape, e.dtype, e.dataId, s), this.trainable = t, this.name = n;
  }
  assign(e) {
    if (e.dtype !== this.dtype)
      throw new Error(`dtype of the new value (${e.dtype}) and previous value (${this.dtype}) must match`);
    if (!kr(e.shape, this.shape))
      throw new Error(`shape of the new value (${e.shape}) and previous value (${this.shape}) must match`);
    cs().disposeTensor(this), this.dataId = e.dataId, cs().incRef(this, null);
  }
  dispose() {
    cs().disposeVariable(this), this.isDisposedInternal = true;
  }
};
Object.defineProperty(xd, Symbol.hasInstance, { value: (e) => e instanceof et && e.assign != null && e.assign instanceof Function });
var _s = {};
Ee(_s, { assertTypesMatch: () => yk, getTensorsInContainer: () => Lg, isTensorInList: () => c_, makeTypesMatch: () => vt });
var u_ = ((e) => (e.R0 = "R0", e.R1 = "R1", e.R2 = "R2", e.R3 = "R3", e.R4 = "R4", e.R5 = "R5", e.R6 = "R6", e))(u_ || {});
var fk = ((e) => (e.float32 = "float32", e.int32 = "int32", e.bool = "int32", e.complex64 = "complex64", e))(fk || {});
var mk = ((e) => (e.float32 = "float32", e.int32 = "int32", e.bool = "bool", e.complex64 = "complex64", e))(mk || {});
var gk = ((e) => (e.float32 = "float32", e.int32 = "float32", e.bool = "float32", e.complex64 = "complex64", e))(gk || {});
var bk = ((e) => (e.float32 = "complex64", e.int32 = "complex64", e.bool = "complex64", e.complex64 = "complex64", e))(bk || {});
var l_ = { float32: gk, int32: fk, bool: mk, complex64: bk };
function cn(e, t) {
  if (e === "string" || t === "string") {
    if (e === "string" && t === "string")
      return "string";
    throw new Error(`Can not upcast ${e} with ${t}`);
  }
  return l_[e][t];
}
function bp(e) {
  return cn(e, "int32");
}
function vt(e, t) {
  if (e.dtype === t.dtype)
    return [e, t];
  let n = cn(e.dtype, t.dtype);
  return [e.cast(n), t.cast(n)];
}
function yk(e, t) {
  F(e.dtype === t.dtype, () => `The dtypes of the first(${e.dtype}) and second(${t.dtype}) input must match`);
}
function c_(e, t) {
  return t.some((n) => n.id === e.id);
}
function Lg(e) {
  let t = [];
  return vk(e, t, /* @__PURE__ */ new Set()), t;
}
function vk(e, t, n) {
  if (e == null)
    return;
  if (e instanceof et) {
    t.push(e);
    return;
  }
  if (!d_(e))
    return;
  let s = e;
  for (let r in s) {
    let a = s[r];
    n.has(a) || (n.add(a), vk(a, t, n));
  }
}
function d_(e) {
  return Array.isArray(e) || typeof e == "object";
}
function Wf(e) {
  return e.kernelName != null;
}
var hx = class {
  constructor() {
    this.registeredVariables = {}, this.nextTapeNodeId = 0, this.numBytes = 0, this.numTensors = 0, this.numStringTensors = 0, this.numDataBuffers = 0, this.gradientDepth = 0, this.kernelDepth = 0, this.scopeStack = [], this.numDataMovesStack = [], this.nextScopeId = 0, this.tensorInfo = /* @__PURE__ */ new WeakMap(), this.profiling = false, this.activeProfile = { newBytes: 0, newTensors: 0, peakBytes: 0, kernels: [], result: null, get kernelNames() {
      return Array.from(new Set(this.kernels.map((e) => e.name)));
    } };
  }
  dispose() {
    for (let e in this.registeredVariables)
      this.registeredVariables[e].dispose();
  }
};
var om = class {
  constructor(e) {
    this.ENV = e, this.registry = {}, this.registryFactory = {}, this.pendingBackendInitId = 0, this.state = new hx();
  }
  async ready() {
    if (this.pendingBackendInit != null)
      return this.pendingBackendInit.then(() => {
      });
    if (this.backendInstance != null)
      return;
    let e = this.getSortedBackends();
    for (let t = 0; t < e.length; t++) {
      let n = e[t];
      if (await this.initializeBackend(n).success) {
        await this.setBackend(n);
        return;
      }
    }
    throw new Error("Could not initialize any backends, all backend initializations failed.");
  }
  get backend() {
    if (this.pendingBackendInit != null)
      throw new Error(`Backend '${this.backendName}' has not yet been initialized. Make sure to await tf.ready() or await tf.setBackend() before calling other methods`);
    if (this.backendInstance == null) {
      let { name: e, asyncInit: t } = this.initializeBackendsAndReturnBest();
      if (t)
        throw new Error(`The highest priority backend '${e}' has not yet been initialized. Make sure to await tf.ready() or await tf.setBackend() before calling other methods`);
      this.setBackend(e);
    }
    return this.backendInstance;
  }
  backendNames() {
    return Object.keys(this.registryFactory);
  }
  findBackend(e) {
    if (!(e in this.registry))
      if (e in this.registryFactory) {
        let { asyncInit: t } = this.initializeBackend(e);
        if (t)
          return null;
      } else
        return null;
    return this.registry[e];
  }
  findBackendFactory(e) {
    return e in this.registryFactory ? this.registryFactory[e].factory : null;
  }
  registerBackend(e, t, n = 1) {
    return e in this.registryFactory ? (rr(`${e} backend was already registered. Reusing existing backend factory.`), false) : (this.registryFactory[e] = { factory: t, priority: n }, true);
  }
  async setBackend(e) {
    if (this.registryFactory[e] == null)
      throw new Error(`Backend name '${e}' not found in registry`);
    if (this.backendName = e, this.registry[e] == null) {
      this.backendInstance = null;
      let { success: t, asyncInit: n } = this.initializeBackend(e);
      if (!(n ? await t : t))
        return false;
    }
    return this.backendInstance = this.registry[e], this.setupRegisteredKernels(), this.profiler = new Y$(this.backendInstance), true;
  }
  setupRegisteredKernels() {
    am(this.backendName).forEach((t) => {
      t.setupFunc != null && t.setupFunc(this.backendInstance);
    });
  }
  disposeRegisteredKernels(e) {
    am(e).forEach((n) => {
      n.disposeFunc != null && n.disposeFunc(this.registry[e]);
    });
  }
  initializeBackend(e) {
    let t = this.registryFactory[e];
    if (t == null)
      throw new Error(`Cannot initialize backend ${e}, no registration found.`);
    try {
      let n = t.factory();
      if (n && !(n instanceof il) && typeof n.then == "function") {
        let s = ++this.pendingBackendInitId, r = n.then((a) => s < this.pendingBackendInitId ? false : (this.registry[e] = a, this.pendingBackendInit = null, true)).catch((a) => (s < this.pendingBackendInitId || (this.pendingBackendInit = null, rr(`Initialization of backend ${e} failed`), rr(a.stack || a.message)), false));
        return this.pendingBackendInit = r, { success: r, asyncInit: true };
      } else
        return this.registry[e] = n, { success: true, asyncInit: false };
    } catch (n) {
      return rr(`Initialization of backend ${e} failed`), rr(n.stack || n.message), { success: false, asyncInit: false };
    }
  }
  removeBackend(e) {
    if (!(e in this.registryFactory))
      throw new Error(`${e} backend not found in registry`);
    this.backendName === e && this.pendingBackendInit != null && this.pendingBackendInitId++, e in this.registry && (this.disposeRegisteredKernels(e), this.registry[e].dispose(), delete this.registry[e]), delete this.registryFactory[e], this.backendName === e && (this.pendingBackendInit = null, this.backendName = null, this.backendInstance = null);
  }
  getSortedBackends() {
    if (Object.keys(this.registryFactory).length === 0)
      throw new Error("No backend found in registry.");
    return Object.keys(this.registryFactory).sort((e, t) => this.registryFactory[t].priority - this.registryFactory[e].priority);
  }
  initializeBackendsAndReturnBest() {
    let e = this.getSortedBackends();
    for (let t = 0; t < e.length; t++) {
      let n = e[t], { success: s, asyncInit: r } = this.initializeBackend(n);
      if (r || s)
        return { name: n, asyncInit: r };
    }
    throw new Error("Could not initialize any backends, all backend initializations failed.");
  }
  moveData(e, t) {
    let n = this.state.tensorInfo.get(t), s = n.backend, r = this.readSync(t), a = s.refCount(t);
    s.disposeData(t, true), n.backend = e, e.move(t, r, n.shape, n.dtype, a), this.shouldCheckForMemLeaks() && this.state.numDataMovesStack[this.state.numDataMovesStack.length - 1]++;
  }
  tidy(e, t) {
    let n = null;
    if (t == null) {
      if (typeof e != "function")
        throw new Error("Please provide a function to tidy()");
      t = e;
    } else {
      if (typeof e != "string" && !(e instanceof String))
        throw new Error("When calling with two arguments, the first argument to tidy() must be a string");
      if (typeof t != "function")
        throw new Error("When calling with two arguments, the 2nd argument to tidy() must be a function");
      n = e;
    }
    let s;
    return this.scopedRun(() => this.startScope(n), () => this.endScope(s), () => (s = t(), s instanceof Promise && console.error("Cannot return a Promise inside of tidy."), s));
  }
  scopedRun(e, t, n) {
    e();
    try {
      let s = n();
      return t(), s;
    } catch (s) {
      throw t(), s;
    }
  }
  nextTensorId() {
    return om.nextTensorId++;
  }
  nextVariableId() {
    return om.nextVariableId++;
  }
  clone(e) {
    let t = M.runKernel(Wa, { x: e }), n = { x: e }, s = (a) => ({ x: () => {
      let i = "float32", o = { x: a }, u = { dtype: i };
      return M.runKernel(Ta, o, u);
    } }), r = [];
    return this.addTapeNode(this.state.activeScope.name, n, [t], s, r, {}), t;
  }
  runKernel(e, t, n) {
    if (this.backendName == null && this.backend, !(rm(e, this.backendName) != null))
      throw new Error(`Kernel '${e}' not registered for backend '${this.backendName}'`);
    return this.runKernelFunc({ kernelName: e, inputs: t, attrs: n });
  }
  shouldCheckForMemLeaks() {
    return this.ENV.getBool("IS_TEST");
  }
  checkKernelForMemLeak(e, t, n) {
    let s = this.backend.numDataIds(), r = 0;
    n.forEach((o) => {
      r += o.dtype === "complex64" ? 3 : 1;
    });
    let a = this.state.numDataMovesStack[this.state.numDataMovesStack.length - 1], i = s - t - r - a;
    if (i > 0)
      throw new Error(`Backend '${this.backendName}' has an internal memory leak (${i} data ids) after running '${e}'`);
  }
  runKernelFunc(e) {
    let t, n = [], s = this.isTapeOn(), r = this.state.numBytes, a = this.state.numTensors;
    this.shouldCheckForMemLeaks() && this.state.numDataMovesStack.push(0);
    let i;
    this.backendName == null && this.backend;
    let o, u = Wf(e) ? e.kernelName : this.state.activeScope != null ? this.state.activeScope.name : "";
    if (Wf(e)) {
      let { kernelName: h, inputs: f, attrs: m } = e;
      this.backendName == null && this.backend;
      let g = rm(h, this.backendName);
      F(g != null, () => `Cannot find registered kernel '${h}' for backend '${this.backendName}'`), i = () => {
        let b = this.backend.numDataIds();
        o = g.kernelFunc({ inputs: f, attrs: m, backend: this.backend });
        let y = Array.isArray(o) ? o : [o];
        this.shouldCheckForMemLeaks() && this.checkKernelForMemLeak(h, b, y);
        let v = y.map((x) => x.rank != null ? x : this.makeTensorFromTensorInfo(x));
        if (s) {
          let x = this.getTensorsForGradient(h, f, v);
          n = this.saveTensorsForBackwardMode(x);
        }
        return v;
      };
    } else {
      let { forwardFunc: h } = e, f = (m) => {
        !s || (n = m.map((g) => this.keep(this.clone(g))));
      };
      i = () => {
        let m = this.backend.numDataIds();
        o = this.tidy(() => h(this.backend, f));
        let g = Array.isArray(o) ? o : [o];
        return this.shouldCheckForMemLeaks() && this.checkKernelForMemLeak(u, m, g), g;
      };
    }
    let { inputs: l, attrs: c } = e, p = Wf(e) ? null : e.backwardsFunc, d;
    return this.scopedRun(() => this.state.kernelDepth++, () => this.state.kernelDepth--, () => {
      !this.ENV.getBool("DEBUG") && !this.state.profiling ? t = i() : (d = this.profiler.profileKernel(u, l, () => i()), this.ENV.getBool("DEBUG") && this.profiler.logKernelProfile(d), t = d.outputs);
    }), s && this.addTapeNode(u, l, t, p, n, c), this.state.profiling && this.state.activeProfile.kernels.push({ name: u, bytesAdded: this.state.numBytes - r, totalBytesSnapshot: this.state.numBytes, tensorsAdded: this.state.numTensors - a, totalTensorsSnapshot: this.state.numTensors, inputShapes: Object.keys(l).map((h) => l[h] != null ? l[h].shape : null), outputShapes: t.map((h) => h.shape), kernelTimeMs: d.timeMs, extraInfo: d.extraInfo }), Array.isArray(o) ? t : t[0];
  }
  saveTensorsForBackwardMode(e) {
    return e.map((n) => this.keep(this.clone(n)));
  }
  getTensorsForGradient(e, t, n) {
    let s = lx(e);
    if (s != null) {
      let r = s.inputsToSave || [], a = s.outputsToSave || [], i;
      s.saveAllInputs ? (F(Array.isArray(t), () => "saveAllInputs is true, expected inputs to be an array."), i = Object.keys(t).map((u) => t[u])) : i = r.map((u) => t[u]);
      let o = n.filter((u, l) => a[l]);
      return i.concat(o);
    }
    return [];
  }
  makeTensor(e, t, n, s) {
    if (e == null)
      throw new Error("Values passed to engine.makeTensor() are null");
    n = n || "float32", s = s || this.backend;
    let r = e;
    n === "string" && ar(e[0]) && (r = e.map((o) => Pl(o)));
    let a = s.write(r, t, n), i = new et(t, n, a, this.nextTensorId());
    if (this.trackTensor(i, s), n === "string") {
      let o = this.state.tensorInfo.get(a), u = ak(r);
      this.state.numBytes += u - o.bytes, o.bytes = u;
    }
    return i;
  }
  makeTensorFromTensorInfo(e, t) {
    let { dataId: n, shape: s, dtype: r } = e, a = new et(s, r, n, this.nextTensorId());
    return this.trackTensor(a, t), a;
  }
  makeVariable(e, t = true, n, s) {
    n = n || this.nextVariableId().toString(), s != null && s !== e.dtype && (e = e.cast(s));
    let r = new xd(e, t, n, this.nextTensorId());
    if (this.state.registeredVariables[r.name] != null)
      throw new Error(`Variable with name ${r.name} was already registered`);
    return this.state.registeredVariables[r.name] = r, this.incRef(r, this.backend), r;
  }
  trackTensor(e, t) {
    this.state.numTensors++, e.dtype === "string" && this.state.numStringTensors++;
    let n = 0;
    e.dtype !== "complex64" && e.dtype !== "string" && (n = e.size * tm(e.dtype)), this.state.numBytes += n, this.state.tensorInfo.has(e.dataId) || (this.state.numDataBuffers++, this.state.tensorInfo.set(e.dataId, { backend: t || this.backend, dtype: e.dtype, shape: e.shape, bytes: n })), e instanceof xd || this.track(e);
  }
  incRef(e, t) {
    this.trackTensor(e, t), this.backend.incRef(e.dataId);
  }
  removeDataId(e, t) {
    this.state.tensorInfo.has(e) && this.state.tensorInfo.get(e).backend === t && (this.state.tensorInfo.delete(e), this.state.numDataBuffers--);
  }
  disposeTensor(e) {
    if (!this.state.tensorInfo.has(e.dataId))
      return;
    let t = this.state.tensorInfo.get(e.dataId);
    if (this.state.numTensors--, e.dtype === "string" && (this.state.numStringTensors--, this.state.numBytes -= t.bytes), e.dtype !== "complex64" && e.dtype !== "string") {
      let n = e.size * tm(e.dtype);
      this.state.numBytes -= n;
    }
    t.backend.disposeData(e.dataId) && this.removeDataId(e.dataId, t.backend);
  }
  disposeVariables() {
    for (let e in this.state.registeredVariables) {
      let t = this.state.registeredVariables[e];
      this.disposeVariable(t);
    }
  }
  disposeVariable(e) {
    this.disposeTensor(e), this.state.registeredVariables[e.name] != null && delete this.state.registeredVariables[e.name];
  }
  memory() {
    let e = this.backend.memory();
    return e.numTensors = this.state.numTensors, e.numDataBuffers = this.state.numDataBuffers, e.numBytes = this.state.numBytes, this.state.numStringTensors > 0 && (e.unreliable = true, e.reasons == null && (e.reasons = []), e.reasons.push("Memory usage by string tensors is approximate (2 bytes per character)")), e;
  }
  async profile(e) {
    this.state.profiling = true;
    let t = this.state.numBytes, n = this.state.numTensors;
    this.state.activeProfile.kernels = [], this.state.activeProfile.result = await e(), this.state.profiling = false, this.state.activeProfile.peakBytes = Math.max(...this.state.activeProfile.kernels.map((s) => s.totalBytesSnapshot)), this.state.activeProfile.newBytes = this.state.numBytes - t, this.state.activeProfile.newTensors = this.state.numTensors - n;
    for (let s of this.state.activeProfile.kernels)
      s.kernelTimeMs = await s.kernelTimeMs, s.extraInfo = await s.extraInfo;
    return this.state.activeProfile;
  }
  isTapeOn() {
    return this.state.gradientDepth > 0 && this.state.kernelDepth === 0;
  }
  addTapeNode(e, t, n, s, r, a) {
    let i = { id: this.state.nextTapeNodeId++, kernelName: e, inputs: t, outputs: n, saved: r }, o = lx(e);
    o != null && (s = o.gradFunc), s != null && (i.gradient = (u) => (u = u.map((l, c) => {
      if (l == null) {
        let p = n[c], d = Zd(p.size, p.dtype);
        return this.makeTensor(d, p.shape, p.dtype);
      }
      return l;
    }), s(u.length > 1 ? u : u[0], r, a))), this.state.activeTape.push(i);
  }
  keep(e) {
    return e.kept = true, e;
  }
  startTape() {
    this.state.gradientDepth === 0 && (this.state.activeTape = []), this.state.gradientDepth++;
  }
  endTape() {
    this.state.gradientDepth--;
  }
  startScope(e) {
    let t = { track: [], name: "unnamed scope", id: this.state.nextScopeId++ };
    e && (t.name = e), this.state.scopeStack.push(t), this.state.activeScope = t;
  }
  endScope(e) {
    let t = Lg(e), n = new Set(t.map((r) => r.id));
    for (let r = 0; r < this.state.activeScope.track.length; r++) {
      let a = this.state.activeScope.track[r];
      !a.kept && !n.has(a.id) && a.dispose();
    }
    let s = this.state.scopeStack.pop();
    this.state.activeScope = this.state.scopeStack.length === 0 ? null : this.state.scopeStack[this.state.scopeStack.length - 1], t.forEach((r) => {
      !r.kept && r.scopeId === s.id && this.track(r);
    });
  }
  gradients(e, t, n, s = false) {
    if (F(t.length > 0, () => "gradients() received an empty list of xs."), n != null && n.dtype !== "float32")
      throw new Error(`dy must have 'float32' dtype, but has '${n.dtype}'`);
    let r = this.scopedRun(() => this.startTape(), () => this.endTape(), () => this.tidy("forward", e));
    F(r instanceof et, () => "The result y returned by f() must be a tensor.");
    let a = J$(this.state.activeTape, t, r);
    if (!s && a.length === 0 && t.length > 0)
      throw new Error("Cannot compute gradient of y=f(x) with respect to x. Make sure that the f you passed encloses all operations that lead from x to y.");
    return this.tidy("backward", () => {
      let i = {};
      i[r.id] = n == null ? p_(r.shape) : n, e_(i, a, (u) => this.tidy(u), h_);
      let o = t.map((u) => i[u.id]);
      return this.state.gradientDepth === 0 && (this.state.activeTape.forEach((u) => {
        for (let l of u.saved)
          l.dispose();
      }), this.state.activeTape = null), { value: r, grads: o };
    });
  }
  customGrad(e) {
    return F(hr(e), () => "The f passed in customGrad(f) must be a function."), (...t) => {
      F(t.every((i) => i instanceof et), () => "The args passed in customGrad(f)(x1, x2,...) must all be tensors");
      let n, s = {};
      t.forEach((i, o) => {
        s[o] = i;
      });
      let r = (i, o) => (n = e(...t, o), F(n.value instanceof et, () => "The function f passed in customGrad(f) must return an object where `obj.value` is a tensor"), F(hr(n.gradFunc), () => "The function f passed in customGrad(f) must return an object where `obj.gradFunc` is a function."), n.value), a = (i, o) => {
        let u = n.gradFunc(i, o), l = Array.isArray(u) ? u : [u];
        F(l.length === t.length, () => "The function f passed in customGrad(f) must return an object where `obj.gradFunc` is a function that returns the same number of tensors as inputs passed to f(...)."), F(l.every((p) => p instanceof et), () => "The function f passed in customGrad(f) must return an object where `obj.gradFunc` is a function that returns a list of only tensors.");
        let c = {};
        return l.forEach((p, d) => {
          c[d] = () => p;
        }), c;
      };
      return this.runKernelFunc({ forwardFunc: r, backwardsFunc: a, inputs: s });
    };
  }
  readSync(e) {
    return this.state.tensorInfo.get(e).backend.readSync(e);
  }
  read(e) {
    return this.state.tensorInfo.get(e).backend.read(e);
  }
  readToGPU(e, t) {
    return this.state.tensorInfo.get(e).backend.readToGPU(e, t);
  }
  async time(e) {
    let t = ju(), n = await this.backend.time(e);
    return n.wallMs = ju() - t, n;
  }
  track(e) {
    return this.state.activeScope != null && (e.scopeId = this.state.activeScope.id, this.state.activeScope.track.push(e)), e;
  }
  get registeredVariables() {
    return this.state.registeredVariables;
  }
  reset() {
    this.pendingBackendInitId++, this.state.dispose(), this.ENV.reset(), this.state = new hx();
    for (let e in this.registry)
      this.disposeRegisteredKernels(e), this.registry[e].dispose(), delete this.registry[e];
    this.backendName = null, this.backendInstance = null, this.pendingBackendInit = null;
  }
};
var Bg = om;
Bg.nextTensorId = 0;
Bg.nextVariableId = 0;
function p_(e) {
  let t = ug(dt(e), "float32");
  return M.makeTensor(t, e, "float32");
}
function xk() {
  let e = ck();
  if (e._tfengine == null) {
    let t = new R$(e);
    e._tfengine = new Bg(t);
  }
  return P$(e._tfengine.ENV), r_(() => e._tfengine), e._tfengine;
}
var M = xk();
function h_(e, t) {
  let n = { a: e, b: t };
  return M.runKernel(Sr, n);
}
var yp = {};
Ee(yp, { isBrowser: () => wk, isMobile: () => g_, mockIsMobile: () => m_ });
function f_() {
  return typeof navigator != "undefined" && navigator != null;
}
var um;
function m_(e) {
  um = e;
}
function g_(e) {
  if (um !== void 0)
    return um;
  if (e || f_()) {
    if (e || (e = navigator), e.product === "ReactNative")
      return true;
    let t = e.userAgent || e.vendor || (typeof window != "undefined" ? window.opera : "");
    if (!t) {
      let n = e;
      return n.userAgentData && n.userAgentData.mobile;
    }
    return /(android|bb\d+|meego).+mobile|avantgo|bada\/|blackberry|blazer|compal|elaine|fennec|hiptop|iemobile|ip(hone|od)|iris|kindle|lge |maemo|midp|mmp|mobile.+firefox|netfront|opera m(ob|in)i|palm( os)?|phone|p(ixi|re)\/|plucker|pocket|psp|series(4|6)0|symbian|treo|up\.(browser|link)|vodafone|wap|windows ce|xda|xiino/i.test(t) || /1207|6310|6590|3gso|4thp|50[1-6]i|770s|802s|a wa|abac|ac(er|oo|s\-)|ai(ko|rn)|al(av|ca|co)|amoi|an(ex|ny|yw)|aptu|ar(ch|go)|as(te|us)|attw|au(di|\-m|r |s )|avan|be(ck|ll|nq)|bi(lb|rd)|bl(ac|az)|br(e|v)w|bumb|bw\-(n|u)|c55\/|capi|ccwa|cdm\-|cell|chtm|cldc|cmd\-|co(mp|nd)|craw|da(it|ll|ng)|dbte|dc\-s|devi|dica|dmob|do(c|p)o|ds(12|\-d)|el(49|ai)|em(l2|ul)|er(ic|k0)|esl8|ez([4-7]0|os|wa|ze)|fetc|fly(\-|_)|g1 u|g560|gene|gf\-5|g\-mo|go(\.w|od)|gr(ad|un)|haie|hcit|hd\-(m|p|t)|hei\-|hi(pt|ta)|hp( i|ip)|hs\-c|ht(c(\-| |_|a|g|p|s|t)|tp)|hu(aw|tc)|i\-(20|go|ma)|i230|iac( |\-|\/)|ibro|idea|ig01|ikom|im1k|inno|ipaq|iris|ja(t|v)a|jbro|jemu|jigs|kddi|keji|kgt( |\/)|klon|kpt |kwc\-|kyo(c|k)|le(no|xi)|lg( g|\/(k|l|u)|50|54|\-[a-w])|libw|lynx|m1\-w|m3ga|m50\/|ma(te|ui|xo)|mc(01|21|ca)|m\-cr|me(rc|ri)|mi(o8|oa|ts)|mmef|mo(01|02|bi|de|do|t(\-| |o|v)|zz)|mt(50|p1|v )|mwbp|mywa|n10[0-2]|n20[2-3]|n30(0|2)|n50(0|2|5)|n7(0(0|1)|10)|ne((c|m)\-|on|tf|wf|wg|wt)|nok(6|i)|nzph|o2im|op(ti|wv)|oran|owg1|p800|pan(a|d|t)|pdxg|pg(13|\-([1-8]|c))|phil|pire|pl(ay|uc)|pn\-2|po(ck|rt|se)|prox|psio|pt\-g|qa\-a|qc(07|12|21|32|60|\-[2-7]|i\-)|qtek|r380|r600|raks|rim9|ro(ve|zo)|s55\/|sa(ge|ma|mm|ms|ny|va)|sc(01|h\-|oo|p\-)|sdk\/|se(c(\-|0|1)|47|mc|nd|ri)|sgh\-|shar|sie(\-|m)|sk\-0|sl(45|id)|sm(al|ar|b3|it|t5)|so(ft|ny)|sp(01|h\-|v\-|v )|sy(01|mb)|t2(18|50)|t6(00|10|18)|ta(gt|lk)|tcl\-|tdg\-|tel(i|m)|tim\-|t\-mo|to(pl|sh)|ts(70|m\-|m3|m5)|tx\-9|up(\.b|g1|si)|utst|v400|v750|veri|vi(rg|te)|vk(40|5[0-3]|\-v)|vm40|voda|vulc|vx(52|53|60|61|70|80|81|83|85|98)|w3c(\-| )|webc|whit|wi(g |nc|nw)|wmlb|wonu|x700|yas\-|your|zeto|zte\-/i.test(t.substr(0, 4));
  }
  return false;
}
function wk() {
  return typeof window != "undefined" && window.document != null || typeof WorkerGlobalScope != "undefined";
}
var Kn = K();
Kn.registerFlag("DEBUG", () => false, (e) => {
  e && console.warn("Debugging mode is ON. The output of every math call will be downloaded to CPU and checked for NaNs. This significantly impacts performance.");
});
Kn.registerFlag("IS_BROWSER", () => wk());
Kn.registerFlag("IS_NODE", () => typeof process != "undefined" && typeof process.versions != "undefined" && typeof process.versions.node != "undefined");
Kn.registerFlag("IS_CHROME", () => typeof navigator != "undefined" && navigator != null && navigator.userAgent != null && /Chrome/.test(navigator.userAgent) && /Google Inc/.test(navigator.vendor));
Kn.registerFlag("PROD", () => false);
Kn.registerFlag("TENSORLIKE_CHECK_SHAPE_CONSISTENCY", () => Kn.getBool("DEBUG"));
Kn.registerFlag("DEPRECATION_WARNINGS_ENABLED", () => true);
Kn.registerFlag("IS_TEST", () => false);
Kn.registerFlag("CHECK_COMPUTATION_FOR_ERRORS", () => true);
Kn.registerFlag("WRAP_TO_IMAGEBITMAP", () => false);
Kn.registerFlag("ENGINE_COMPILE_ONLY", () => false);
function Rs(e, t) {
  let n = e;
  if (Qt(e))
    return t === "string" ? [] : [e.length];
  if (!Array.isArray(e))
    return [];
  let s = [];
  for (; Array.isArray(n) || Qt(n) && t !== "string"; )
    s.push(n.length), n = n[0];
  return Array.isArray(e) && K().getBool("TENSORLIKE_CHECK_SHAPE_CONSISTENCY") && kk(e, s, []), s;
}
function kk(e, t, n) {
  if (n = n || [], !Array.isArray(e) && !Qt(e)) {
    F(t.length === 0, () => `Element arr[${n.join("][")}] is a primitive, but should be an array/TypedArray of ${t[0]} elements`);
    return;
  }
  F(t.length > 0, () => `Element arr[${n.join("][")}] should be a primitive, but is an array of ${e.length} elements`), F(e.length === t[0], () => `Element arr[${n.join("][")}] should have ${t[0]} elements, but has ${e.length} elements`);
  let s = t.slice(1);
  for (let r = 0; r < e.length; ++r)
    kk(e[r], s, n.concat(r));
}
function fx(e, t, n, s) {
  if (e !== "string_or_numeric") {
    if (e == null)
      throw new Error("Expected dtype cannot be null.");
    if (e !== "numeric" && e !== t || e === "numeric" && t === "string")
      throw new Error(`Argument '${n}' passed to '${s}' must be ${e} tensor, but got ${t} tensor`);
  }
}
function _(e, t, n, s = "numeric") {
  if (e instanceof et)
    return fx(s, e.dtype, t, n), e;
  let r = Qd(e);
  if (r !== "string" && ["bool", "int32", "float32"].indexOf(s) >= 0 && (r = s), fx(s, r, t, n), e == null || !Qt(e) && !Array.isArray(e) && typeof e != "number" && typeof e != "boolean" && typeof e != "string") {
    let u = e == null ? "null" : e.constructor.name;
    throw new Error(`Argument '${t}' passed to '${n}' must be a Tensor or TensorLike, but got '${u}'`);
  }
  let a = Rs(e, r);
  !Qt(e) && !Array.isArray(e) && (e = [e]);
  let o = r !== "string" ? gp(e, r) : ra(e, [], true);
  return M.makeTensor(o, a, r);
}
function Ku(e, t, n, s = "numeric") {
  if (!Array.isArray(e))
    throw new Error(`Argument ${t} passed to ${n} must be a \`Tensor[]\` or \`TensorLike[]\``);
  return e.map((a, i) => _(a, `${t}[${i}]`, n, s));
}
var b_ = "__op";
function L(e) {
  let t = Object.keys(e);
  if (t.length !== 1)
    throw new Error(`Please provide an object with a single key (operation name) mapping to a function. Got an object with ${t.length} keys.`);
  let n = t[0], s = e[n];
  n.endsWith("_") && (n = n.substring(0, n.length - 1)), n = n + b_;
  let r = (...a) => {
    M.startScope(n);
    try {
      let i = s(...a);
      return cg(i) && console.error("Cannot return a Promise inside of tidy."), M.endScope(i), i;
    } catch (i) {
      throw M.endScope(null), i;
    }
  };
  return Object.defineProperty(r, "name", { value: n, configurable: true }), r;
}
function y_(e, t) {
  let n = _(e, "real", "complex"), s = _(t, "imag", "complex");
  pn(n.shape, s.shape, `real and imag shapes, ${n.shape} and ${s.shape}, must match in call to tf.complex().`);
  let r = { real: n, imag: s };
  return M.runKernel(ep, r);
}
var ua = L({ complex_: y_ });
function Nr(e, t, n, s) {
  if (s == null && (s = Qd(e)), s === "complex64")
    throw new Error("Cannot construct a complex64 tensor directly. Please use tf.complex(real, imag).");
  if (!Qt(e) && !Array.isArray(e) && typeof e != "number" && typeof e != "boolean" && typeof e != "string")
    throw new Error("values passed to tensor(values) must be a number/boolean/string or an array of numbers/booleans/strings, or a TypedArray");
  if (t != null) {
    lg(t);
    let r = dt(t), a = dt(n);
    F(r === a, () => `Based on the provided shape, [${t}], the tensor should have ${r} values but has ${a}`);
    for (let i = 0; i < n.length; ++i) {
      let o = n[i], u = i === n.length - 1 ? o !== dt(t.slice(i)) : true;
      F(n[i] === t[i] || !u, () => `Error creating a new Tensor. Inferred shape (${n}) does not match the provided shape (${t}). `);
    }
  }
  return !Qt(e) && !Array.isArray(e) && (e = [e]), t = t || n, e = s !== "string" ? gp(e, s) : ra(e, [], true), M.makeTensor(e, t, s);
}
function ms(e, t, n) {
  let s = Rs(e, n);
  return Nr(e, t, s, n);
}
var lm = { float32: 4, float16: 2, int32: 4, uint16: 2, uint8: 1, bool: 1, complex64: 8 };
var wd = 4;
async function v_(e, t) {
  let n = [], s = [], r = Array.isArray(e) ? e.map((i) => i.name) : Object.keys(e);
  for (let i = 0; i < r.length; ++i) {
    let o = r[i], u = Array.isArray(e) ? e[i].tensor : e[o];
    if (u.dtype !== "float32" && u.dtype !== "int32" && u.dtype !== "bool" && u.dtype !== "string" && u.dtype !== "complex64")
      throw new Error(`Unsupported dtype in weight '${o}': ${u.dtype}`);
    let l = { name: o, shape: u.shape, dtype: u.dtype };
    if (u.dtype === "string") {
      let c = new Promise(async (p) => {
        let d = await u.bytes(), h = d.reduce((g, b) => g + b.length, 0) + wd * d.length, f = new Uint8Array(h), m = 0;
        for (let g = 0; g < d.length; g++) {
          let b = d[g], y = new Uint8Array(new Uint32Array([b.length]).buffer);
          f.set(y, m), m += wd, f.set(b, m), m += b.length;
        }
        p(f);
      });
      s.push(c);
    } else
      s.push(u.data());
    t != null && (l.group = t), n.push(l);
  }
  let a = await Promise.all(s);
  return { data: x_(a), specs: n };
}
function Sk(e, t) {
  let n = {}, s, r = 0;
  for (let a of t) {
    let i = a.name, o = a.dtype, u = a.shape, l = dt(u), c;
    if ("quantization" in a) {
      let p = a.quantization;
      if (p.dtype === "uint8" || p.dtype === "uint16") {
        if (!("min" in p && "scale" in p))
          throw new Error(`Weight ${a.name} with quantization ${p.dtype} doesn't have corresponding metadata min and scale.`);
      } else if (p.dtype === "float16") {
        if (o !== "float32")
          throw new Error(`Weight ${a.name} is quantized with ${p.dtype} which only supports weights of type float32 not ${o}.`);
      } else
        throw new Error(`Weight ${a.name} has unknown quantization dtype ${p.dtype}. Supported quantization dtypes are: 'uint8', 'uint16', and 'float16'.`);
      let d = lm[p.dtype], h = e.slice(r, r + l * d), f = p.dtype === "uint8" ? new Uint8Array(h) : new Uint16Array(h);
      if (o === "float32")
        if (p.dtype === "uint8" || p.dtype === "uint16") {
          c = new Float32Array(f.length);
          for (let m = 0; m < f.length; m++) {
            let g = f[m];
            c[m] = g * p.scale + p.min;
          }
        } else if (p.dtype === "float16")
          s === void 0 && (s = N_()), c = s(f);
        else
          throw new Error(`Unsupported quantization type ${p.dtype} for weight type float32.`);
      else if (o === "int32") {
        if (p.dtype !== "uint8" && p.dtype !== "uint16")
          throw new Error(`Unsupported quantization type ${p.dtype} for weight type int32.`);
        c = new Int32Array(f.length);
        for (let m = 0; m < f.length; m++) {
          let g = f[m];
          c[m] = Math.round(g * p.scale + p.min);
        }
      } else
        throw new Error(`Unsupported dtype in weight '${i}': ${o}`);
      r += l * d;
    } else if (o === "string") {
      let p = dt(a.shape);
      c = [];
      for (let d = 0; d < p; d++) {
        let h = new Uint32Array(e.slice(r, r + wd))[0];
        r += wd;
        let f = new Uint8Array(e.slice(r, r + h));
        c.push(f), r += h;
      }
    } else {
      let p = lm[o], d = e.slice(r, r + l * p);
      if (o === "float32")
        c = new Float32Array(d);
      else if (o === "int32")
        c = new Int32Array(d);
      else if (o === "bool")
        c = new Uint8Array(d);
      else if (o === "complex64") {
        c = new Float32Array(d);
        let h = new Float32Array(c.length / 2), f = new Float32Array(c.length / 2);
        for (let b = 0; b < h.length; b++)
          h[b] = c[b * 2], f[b] = c[b * 2 + 1];
        let m = ms(h, u, "float32"), g = ms(f, u, "float32");
        n[i] = ua(m, g), m.dispose(), g.dispose();
      } else
        throw new Error(`Unsupported dtype in weight '${i}': ${o}`);
      r += l * p;
    }
    o !== "complex64" && (n[i] = ms(c, u, o));
  }
  return n;
}
function x_(e) {
  if (e === null)
    throw new Error(`Invalid input value: ${JSON.stringify(e)}`);
  let t = 0, n = [];
  e.forEach((a) => {
    if (t += a.byteLength, n.push(a.byteLength === a.buffer.byteLength ? a : new a.constructor(a)), !(a instanceof Float32Array || a instanceof Int32Array || a instanceof Uint8Array))
      throw new Error(`Unsupported TypedArray subtype: ${a.constructor.name}`);
  });
  let s = new Uint8Array(t), r = 0;
  return n.forEach((a) => {
    s.set(new Uint8Array(a.buffer), r), r += a.byteLength;
  }), s.buffer;
}
var Vg = typeof Buffer != "undefined" && (typeof Blob == "undefined" || typeof atob == "undefined" || typeof btoa == "undefined");
function mx(e) {
  return Vg ? Buffer.byteLength(e) : new Blob([e]).size;
}
function w_(e) {
  if (Vg)
    return Buffer.from(e).toString("base64");
  let t = new Uint8Array(e), n = "";
  for (let s = 0, r = t.length; s < r; s++)
    n += String.fromCharCode(t[s]);
  return btoa(n);
}
function k_(e) {
  if (Vg) {
    let s = Buffer.from(e, "base64");
    return s.buffer.slice(s.byteOffset, s.byteOffset + s.byteLength);
  }
  let t = atob(e), n = new Uint8Array(t.length);
  for (let s = 0; s < t.length; ++s)
    n.set([t.charCodeAt(s)], s);
  return n.buffer;
}
function Wg(e) {
  if (e.length === 1)
    return e[0];
  let t = 0;
  e.forEach((r) => {
    t += r.byteLength;
  });
  let n = new Uint8Array(t), s = 0;
  return e.forEach((r) => {
    n.set(new Uint8Array(r), s), s += r.byteLength;
  }), n.buffer;
}
function gx(e) {
  let t = "/";
  for (e = e.trim(); e.endsWith(t); )
    e = e.slice(0, e.length - 1);
  let n = e.split(t);
  return n[n.length - 1];
}
function Ik(e, t) {
  let n = { modelTopology: e.modelTopology, format: e.format, generatedBy: e.generatedBy, convertedBy: e.convertedBy, weightsManifest: t };
  return e.signature != null && (n.signature = e.signature), e.userDefinedMetadata != null && (n.userDefinedMetadata = e.userDefinedMetadata), e.modelInitializer != null && (n.modelInitializer = e.modelInitializer), e.trainingConfig != null && (n.trainingConfig = e.trainingConfig), n;
}
async function Ug(e, t) {
  let n = { modelTopology: e.modelTopology, format: e.format, generatedBy: e.generatedBy, convertedBy: e.convertedBy };
  if (e.trainingConfig != null && (n.trainingConfig = e.trainingConfig), e.weightsManifest != null) {
    let [s, r] = await t(e.weightsManifest);
    n.weightSpecs = s, n.weightData = r;
  }
  return e.signature != null && (n.signature = e.signature), e.userDefinedMetadata != null && (n.userDefinedMetadata = e.userDefinedMetadata), e.modelInitializer != null && (n.modelInitializer = e.modelInitializer), n;
}
function zl(e) {
  if (e.modelTopology instanceof ArrayBuffer)
    throw new Error("Expected JSON model topology, received ArrayBuffer.");
  return { dateSaved: new Date(), modelTopologyType: "JSON", modelTopologyBytes: e.modelTopology == null ? 0 : mx(JSON.stringify(e.modelTopology)), weightSpecsBytes: e.weightSpecs == null ? 0 : mx(JSON.stringify(e.weightSpecs)), weightDataBytes: e.weightData == null ? 0 : e.weightData.byteLength };
}
function S_() {
  let e = (n) => {
    let s = n << 13, r = 0;
    for (; (s & 8388608) === 0; )
      r -= 8388608, s <<= 1;
    return s &= -8388609, r += 947912704, s | r;
  }, t = new Uint32Array(2048);
  t[0] = 0;
  for (let n = 1; n < 1024; n++)
    t[n] = e(n);
  for (let n = 1024; n < 2048; n++)
    t[n] = 939524096 + (n - 1024 << 13);
  return t;
}
function I_() {
  let e = new Uint32Array(64);
  e[0] = 0, e[31] = 1199570944, e[32] = 2147483648, e[63] = 3347054592;
  for (let t = 1; t < 31; t++)
    e[t] = t << 23;
  for (let t = 33; t < 63; t++)
    e[t] = 2147483648 + (t - 32 << 23);
  return e;
}
function C_() {
  let e = new Uint32Array(64);
  for (let t = 0; t < 64; t++)
    e[t] = 1024;
  return e[0] = e[32] = 0, e;
}
function N_() {
  let e = S_(), t = I_(), n = C_();
  return (s) => {
    let r = new ArrayBuffer(4 * s.length), a = new Uint32Array(r);
    for (let i = 0; i < s.length; i++) {
      let o = s[i], u = e[n[o >> 10] + (o & 1023)] + t[o >> 10];
      a[i] = u;
    }
    return new Float32Array(r);
  };
}
var xt = class {
  constructor() {
    this.saveRouters = [], this.loadRouters = [];
  }
  static getInstance() {
    return xt.instance == null && (xt.instance = new xt()), xt.instance;
  }
  static registerSaveRouter(e) {
    xt.getInstance().saveRouters.push(e);
  }
  static registerLoadRouter(e) {
    xt.getInstance().loadRouters.push(e);
  }
  static getSaveHandlers(e) {
    return xt.getHandlers(e, "save");
  }
  static getLoadHandlers(e, t) {
    return xt.getHandlers(e, "load", t);
  }
  static getHandlers(e, t, n) {
    let s = [];
    return (t === "load" ? xt.getInstance().loadRouters : xt.getInstance().saveRouters).forEach((a) => {
      let i = a(e, n);
      i !== null && s.push(i);
    }), s;
  }
};
var T_ = (e) => xt.registerSaveRouter(e);
var $_ = (e) => xt.registerLoadRouter(e);
var __ = (e) => xt.getSaveHandlers(e);
var A_ = (e, t) => xt.getLoadHandlers(e, t);
var cm = "tensorflowjs";
var dm = 1;
var Qr = "models_store";
var ir = "model_info_store";
function Ck() {
  if (!K().getBool("IS_BROWSER"))
    throw new Error("Failed to obtain IndexedDB factory because the current environmentis not a web browser.");
  let e = typeof window == "undefined" ? self : window, t = e.indexedDB || e.mozIndexedDB || e.webkitIndexedDB || e.msIndexedDB || e.shimIndexedDB;
  if (t == null)
    throw new Error("The current browser does not appear to support IndexedDB.");
  return t;
}
function pm(e) {
  let t = e.result;
  t.createObjectStore(Qr, { keyPath: "modelPath" }), t.createObjectStore(ir, { keyPath: "modelPath" });
}
var la = class {
  constructor(e) {
    if (this.indexedDB = Ck(), e == null || !e)
      throw new Error("For IndexedDB, modelPath must not be null, undefined or empty.");
    this.modelPath = e;
  }
  async save(e) {
    if (e.modelTopology instanceof ArrayBuffer)
      throw new Error("BrowserLocalStorage.save() does not support saving model topology in binary formats yet.");
    return this.databaseAction(this.modelPath, e);
  }
  async load() {
    return this.databaseAction(this.modelPath);
  }
  databaseAction(e, t) {
    return new Promise((n, s) => {
      let r = this.indexedDB.open(cm, dm);
      r.onupgradeneeded = () => pm(r), r.onsuccess = () => {
        let a = r.result;
        if (t == null) {
          let i = a.transaction(Qr, "readonly"), u = i.objectStore(Qr).get(this.modelPath);
          u.onsuccess = () => {
            if (u.result == null)
              return a.close(), s(new Error(`Cannot find model with path '${this.modelPath}' in IndexedDB.`));
            n(u.result.modelArtifacts);
          }, u.onerror = (l) => (a.close(), s(u.error)), i.oncomplete = () => a.close();
        } else {
          let i = zl(t), o = a.transaction(ir, "readwrite"), u = o.objectStore(ir), l = u.put({ modelPath: this.modelPath, modelArtifactsInfo: i }), c;
          l.onsuccess = () => {
            c = a.transaction(Qr, "readwrite");
            let d = c.objectStore(Qr).put({ modelPath: this.modelPath, modelArtifacts: t, modelArtifactsInfo: i });
            d.onsuccess = () => n({ modelArtifactsInfo: i }), d.onerror = (h) => {
              u = o.objectStore(ir);
              let f = u.delete(this.modelPath);
              f.onsuccess = () => (a.close(), s(d.error)), f.onerror = (m) => (a.close(), s(d.error));
            };
          }, l.onerror = (p) => (a.close(), s(l.error)), o.oncomplete = () => {
            c == null ? a.close() : c.oncomplete = () => a.close();
          };
        }
      }, r.onerror = (a) => s(r.error);
    });
  }
};
la.URL_SCHEME = "indexeddb://";
var Nk = (e) => K().getBool("IS_BROWSER") && !Array.isArray(e) && e.startsWith(la.URL_SCHEME) ? E_(e.slice(la.URL_SCHEME.length)) : null;
xt.registerSaveRouter(Nk);
xt.registerLoadRouter(Nk);
function E_(e) {
  return new la(e);
}
function R_(e) {
  return e.startsWith(la.URL_SCHEME) ? e.slice(la.URL_SCHEME.length) : e;
}
var D_ = class {
  constructor() {
    this.indexedDB = Ck();
  }
  async listModels() {
    return new Promise((e, t) => {
      let n = this.indexedDB.open(cm, dm);
      n.onupgradeneeded = () => pm(n), n.onsuccess = () => {
        let s = n.result, r = s.transaction(ir, "readonly"), i = r.objectStore(ir).getAll();
        i.onsuccess = () => {
          let o = {};
          for (let u of i.result)
            o[u.modelPath] = u.modelArtifactsInfo;
          e(o);
        }, i.onerror = (o) => (s.close(), t(i.error)), r.oncomplete = () => s.close();
      }, n.onerror = (s) => t(n.error);
    });
  }
  async removeModel(e) {
    return e = R_(e), new Promise((t, n) => {
      let s = this.indexedDB.open(cm, dm);
      s.onupgradeneeded = () => pm(s), s.onsuccess = () => {
        let r = s.result, a = r.transaction(ir, "readwrite"), i = a.objectStore(ir), o = i.get(e), u;
        o.onsuccess = () => {
          if (o.result == null)
            return r.close(), n(new Error(`Cannot find model with path '${e}' in IndexedDB.`));
          {
            let l = i.delete(e), c = () => {
              u = r.transaction(Qr, "readwrite");
              let d = u.objectStore(Qr).delete(e);
              d.onsuccess = () => t(o.result.modelArtifactsInfo), d.onerror = (h) => n(o.error);
            };
            l.onsuccess = c, l.onerror = (p) => (c(), r.close(), n(o.error));
          }
        }, o.onerror = (l) => (r.close(), n(o.error)), a.oncomplete = () => {
          u == null ? r.close() : u.oncomplete = () => r.close();
        };
      }, s.onerror = (r) => n(s.error);
    });
  }
};
var Us = "/";
var qi = "tensorflowjs_models";
var Tk = "info";
var F_ = "model_topology";
var O_ = "weight_specs";
var P_ = "weight_data";
var z_ = "model_metadata";
function $k(e) {
  return { info: [qi, e, Tk].join(Us), topology: [qi, e, F_].join(Us), weightSpecs: [qi, e, O_].join(Us), weightData: [qi, e, P_].join(Us), modelMetadata: [qi, e, z_].join(Us) };
}
function _k(e) {
  for (let t of Object.values(e))
    window.localStorage.removeItem(t);
}
function M_(e) {
  let t = e.split(Us);
  if (t.length < 3)
    throw new Error(`Invalid key format: ${e}`);
  return t.slice(1, t.length - 1).join(Us);
}
function L_(e) {
  return e.startsWith(ca.URL_SCHEME) ? e.slice(ca.URL_SCHEME.length) : e;
}
var ca = class {
  constructor(e) {
    if (!K().getBool("IS_BROWSER") || typeof window == "undefined" || typeof window.localStorage == "undefined")
      throw new Error("The current environment does not support local storage.");
    if (this.LS = window.localStorage, e == null || !e)
      throw new Error("For local storage, modelPath must not be null, undefined or empty.");
    this.modelPath = e, this.keys = $k(this.modelPath);
  }
  async save(e) {
    if (e.modelTopology instanceof ArrayBuffer)
      throw new Error("BrowserLocalStorage.save() does not support saving model topology in binary formats yet.");
    {
      let t = JSON.stringify(e.modelTopology), n = JSON.stringify(e.weightSpecs), s = zl(e);
      try {
        this.LS.setItem(this.keys.info, JSON.stringify(s)), this.LS.setItem(this.keys.topology, t), this.LS.setItem(this.keys.weightSpecs, n), this.LS.setItem(this.keys.weightData, w_(e.weightData));
        let r = { format: e.format, generatedBy: e.generatedBy, convertedBy: e.convertedBy, signature: e.signature != null ? e.signature : void 0, userDefinedMetadata: e.userDefinedMetadata != null ? e.userDefinedMetadata : void 0, modelInitializer: e.modelInitializer != null ? e.modelInitializer : void 0, trainingConfig: e.trainingConfig != null ? e.trainingConfig : void 0 };
        return this.LS.setItem(this.keys.modelMetadata, JSON.stringify(r)), { modelArtifactsInfo: s };
      } catch (r) {
        throw _k(this.keys), new Error(`Failed to save model '${this.modelPath}' to local storage: size quota being exceeded is a possible cause of this failure: modelTopologyBytes=${s.modelTopologyBytes}, weightSpecsBytes=${s.weightSpecsBytes}, weightDataBytes=${s.weightDataBytes}.`);
      }
    }
  }
  async load() {
    let e = JSON.parse(this.LS.getItem(this.keys.info));
    if (e == null)
      throw new Error(`In local storage, there is no model with name '${this.modelPath}'`);
    if (e.modelTopologyType !== "JSON")
      throw new Error("BrowserLocalStorage does not support loading non-JSON model topology yet.");
    let t = {}, n = JSON.parse(this.LS.getItem(this.keys.topology));
    if (n == null)
      throw new Error(`In local storage, the topology of model '${this.modelPath}' is missing.`);
    t.modelTopology = n;
    let s = JSON.parse(this.LS.getItem(this.keys.weightSpecs));
    if (s == null)
      throw new Error(`In local storage, the weight specs of model '${this.modelPath}' are missing.`);
    t.weightSpecs = s;
    let r = this.LS.getItem(this.keys.modelMetadata);
    if (r != null) {
      let i = JSON.parse(r);
      t.format = i.format, t.generatedBy = i.generatedBy, t.convertedBy = i.convertedBy, i.signature != null && (t.signature = i.signature), i.userDefinedMetadata != null && (t.userDefinedMetadata = i.userDefinedMetadata), i.modelInitializer != null && (t.modelInitializer = i.modelInitializer), i.trainingConfig != null && (t.trainingConfig = i.trainingConfig);
    }
    let a = this.LS.getItem(this.keys.weightData);
    if (a == null)
      throw new Error(`In local storage, the binary weight values of model '${this.modelPath}' are missing.`);
    return t.weightData = k_(a), t;
  }
};
ca.URL_SCHEME = "localstorage://";
var Ak = (e) => K().getBool("IS_BROWSER") && !Array.isArray(e) && e.startsWith(ca.URL_SCHEME) ? B_(e.slice(ca.URL_SCHEME.length)) : null;
xt.registerSaveRouter(Ak);
xt.registerLoadRouter(Ak);
function B_(e) {
  return new ca(e);
}
var V_ = class {
  constructor() {
    F(K().getBool("IS_BROWSER"), () => "Current environment is not a web browser"), F(typeof window == "undefined" || typeof window.localStorage != "undefined", () => "Current browser does not appear to support localStorage"), this.LS = window.localStorage;
  }
  async listModels() {
    let e = {}, t = qi + Us, n = Us + Tk;
    for (let s = 0; s < this.LS.length; ++s) {
      let r = this.LS.key(s);
      if (r.startsWith(t) && r.endsWith(n)) {
        let a = M_(r);
        e[a] = JSON.parse(this.LS.getItem(r));
      }
    }
    return e;
  }
  async removeModel(e) {
    e = L_(e);
    let t = $k(e);
    if (this.LS.getItem(t.info) == null)
      throw new Error(`Cannot find model at path '${e}'`);
    let n = JSON.parse(this.LS.getItem(t.info));
    return _k(t), n;
  }
};
var Yi = "://";
var zn = class {
  constructor() {
    this.managers = {};
  }
  static getInstance() {
    return zn.instance == null && (zn.instance = new zn()), zn.instance;
  }
  static registerManager(e, t) {
    F(e != null, () => "scheme must not be undefined or null."), e.endsWith(Yi) && (e = e.slice(0, e.indexOf(Yi))), F(e.length > 0, () => "scheme must not be an empty string.");
    let n = zn.getInstance();
    F(n.managers[e] == null, () => `A model store manager is already registered for scheme '${e}'.`), n.managers[e] = t;
  }
  static getManager(e) {
    let t = this.getInstance().managers[e];
    if (t == null)
      throw new Error(`Cannot find model manager for scheme '${e}'`);
    return t;
  }
  static getSchemes() {
    return Object.keys(this.getInstance().managers);
  }
};
function rd(e) {
  if (e.indexOf(Yi) === -1)
    throw new Error(`The url string provided does not contain a scheme. Supported schemes are: ${zn.getSchemes().join(",")}`);
  return { scheme: e.split(Yi)[0], path: e.split(Yi)[1] };
}
async function Ek(e, t, n = false) {
  F(e !== t, () => `Old path and new path are the same: '${e}'`);
  let s = xt.getLoadHandlers(e);
  F(s.length > 0, () => `Copying failed because no load handler is found for source URL ${e}.`), F(s.length < 2, () => `Copying failed because more than one (${s.length}) load handlers for source URL ${e}.`);
  let r = s[0], a = xt.getSaveHandlers(t);
  F(a.length > 0, () => `Copying failed because no save handler is found for destination URL ${t}.`), F(a.length < 2, () => `Copying failed because more than one (${s.length}) save handlers for destination URL ${t}.`);
  let i = a[0], o = rd(e).scheme, u = rd(e).path, l = o === rd(e).scheme, c = await r.load();
  n && l && await zn.getManager(o).removeModel(u);
  let p = await i.save(c);
  return n && !l && await zn.getManager(o).removeModel(u), p.modelArtifactsInfo;
}
async function W_() {
  let e = zn.getSchemes(), t = {};
  for (let n of e) {
    let s = await zn.getManager(n).listModels();
    for (let r in s) {
      let a = n + Yi + r;
      t[a] = s[r];
    }
  }
  return t;
}
async function U_(e) {
  let t = rd(e);
  return zn.getManager(t.scheme).removeModel(t.path);
}
async function G_(e, t) {
  return Ek(e, t, false);
}
async function H_(e, t) {
  return Ek(e, t, true);
}
var q_ = class {
  fetch(e, t) {
    return fetch(e, t);
  }
  now() {
    return performance.now();
  }
  encode(e, t) {
    if (t !== "utf-8" && t !== "utf8")
      throw new Error(`Browser's encoder only supports utf-8, but got ${t}`);
    return this.textEncoder == null && (this.textEncoder = new TextEncoder()), this.textEncoder.encode(e);
  }
  decode(e, t) {
    return new TextDecoder(t).decode(e);
  }
};
if (K().get("IS_BROWSER")) {
  K().setPlatform("browser", new q_());
  try {
    zn.registerManager(ca.URL_SCHEME, new V_());
  } catch (e) {
  }
  try {
    zn.registerManager(la.URL_SCHEME, new D_());
  } catch (e) {
  }
}
var j_ = { importFetch: () => e$() };
var Uf;
var K_ = class {
  constructor() {
    this.util = t$(), this.textEncoder = new this.util.TextEncoder();
  }
  fetch(e, t) {
    return K().global.fetch != null ? K().global.fetch(e, t) : (Uf == null && (Uf = j_.importFetch()), Uf(e, t));
  }
  now() {
    let e = process.hrtime();
    return e[0] * 1e3 + e[1] / 1e6;
  }
  encode(e, t) {
    if (t !== "utf-8" && t !== "utf8")
      throw new Error(`Node built-in encoder only supports utf-8, but got ${t}`);
    return this.textEncoder.encode(e);
  }
  decode(e, t) {
    return e.length === 0 ? "" : new this.util.TextDecoder(t).decode(e);
  }
};
K().get("IS_NODE") && !K().get("IS_BROWSER") && K().setPlatform("node", new K_());
function Ae(e, t = "float32", n) {
  return t = t || "float32", lg(e), new Wt(e, t, n);
}
function X_(e, t) {
  let n = _(e, "x", "cast");
  if (!rk(t))
    throw new Error(`Failed to cast to unknown dtype ${t}`);
  if (t === "string" && n.dtype !== "string" || t !== "string" && n.dtype === "string")
    throw new Error("Only strings can be casted to strings");
  let s = { x: n }, r = { dtype: t };
  return M.runKernel(Ta, s, r);
}
var le = L({ cast_: X_ });
function Y_(e) {
  let n = { x: _(e, "x", "clone", "string_or_numeric") };
  return M.runKernel(Wa, n);
}
var ur = L({ clone_: Y_ });
function Q_(e, t = false) {
  console.log(e.toString(t));
}
xk();
var Z_ = { buffer: Ae, cast: le, clone: ur, print: Q_ };
a_(Z_);
var An = {};
Ee(An, { browserFiles: () => aA, browserHTTPRequest: () => cA, concatenateArrayBuffers: () => Wg, copyModel: () => G_, decodeWeights: () => Sk, encodeWeights: () => v_, fromMemory: () => pA, getLoadHandlers: () => A_, getModelArtifactsForJSON: () => Ug, getModelArtifactsInfoForJSON: () => zl, getSaveHandlers: () => __, http: () => Hg, isHTTPScheme: () => fm, listModels: () => W_, loadWeights: () => iA, moveModel: () => H_, registerLoadRouter: () => $_, registerSaveRouter: () => T_, removeModel: () => U_, weightsLoaderFactory: () => Dk, withSaveHandler: () => hA });
var J_ = "model";
var eA = ".json";
var tA = ".weights.bin";
function bx(e) {
  return new Promise((t) => setTimeout(t)).then(e);
}
var hm = class {
  constructor(e) {
    if (!K().getBool("IS_BROWSER"))
      throw new Error("browserDownloads() cannot proceed because the current environment is not a browser.");
    e.startsWith(hm.URL_SCHEME) && (e = e.slice(hm.URL_SCHEME.length)), (e == null || e.length === 0) && (e = J_), this.modelJsonFileName = e + eA, this.weightDataFileName = e + tA;
  }
  async save(e) {
    if (typeof document == "undefined")
      throw new Error("Browser downloads are not supported in this environment since `document` is not present");
    let t = window.URL.createObjectURL(new Blob([e.weightData], { type: "application/octet-stream" }));
    if (e.modelTopology instanceof ArrayBuffer)
      throw new Error("BrowserDownloads.save() does not support saving model topology in binary formats yet.");
    {
      let n = [{ paths: ["./" + this.weightDataFileName], weights: e.weightSpecs }], s = Ik(e, n), r = window.URL.createObjectURL(new Blob([JSON.stringify(s)], { type: "application/json" })), a = this.modelJsonAnchor == null ? document.createElement("a") : this.modelJsonAnchor;
      if (a.download = this.modelJsonFileName, a.href = r, await bx(() => a.dispatchEvent(new MouseEvent("click"))), e.weightData != null) {
        let i = this.weightDataAnchor == null ? document.createElement("a") : this.weightDataAnchor;
        i.download = this.weightDataFileName, i.href = t, await bx(() => i.dispatchEvent(new MouseEvent("click")));
      }
      return { modelArtifactsInfo: zl(e) };
    }
  }
};
var kd = hm;
kd.URL_SCHEME = "downloads://";
var nA = class {
  constructor(e) {
    if (e == null || e.length < 1)
      throw new Error(`When calling browserFiles, at least 1 file is required, but received ${e}`);
    this.jsonFile = e[0], this.weightsFiles = e.slice(1);
  }
  async load() {
    return new Promise((e, t) => {
      let n = new FileReader();
      n.onload = (s) => {
        let r = JSON.parse(s.target.result), a = r.modelTopology;
        if (a == null) {
          t(new Error(`modelTopology field is missing from file ${this.jsonFile.name}`));
          return;
        }
        if (r.weightsManifest == null) {
          t(new Error(`weightManifest field is missing from file ${this.jsonFile.name}`));
          return;
        }
        if (this.weightsFiles.length === 0) {
          e({ modelTopology: a });
          return;
        }
        let o = Ug(r, (u) => this.loadWeights(u));
        e(o);
      }, n.onerror = (s) => t(`Failed to read model topology and weights manifest JSON from file '${this.jsonFile.name}'. BrowserFiles supports loading Keras-style tf.Model artifacts only.`), n.readAsText(this.jsonFile);
    });
  }
  loadWeights(e) {
    let t = [], n = [];
    for (let a of e)
      t.push(...a.weights), n.push(...a.paths);
    let s = this.checkManifestAndWeightFiles(e), r = n.map((a) => this.loadWeightsFile(a, s[a]));
    return Promise.all(r).then((a) => [t, Wg(a)]);
  }
  loadWeightsFile(e, t) {
    return new Promise((n, s) => {
      let r = new FileReader();
      r.onload = (a) => {
        let i = a.target.result;
        n(i);
      }, r.onerror = (a) => s(`Failed to weights data from file of path '${e}'.`), r.readAsArrayBuffer(t);
    });
  }
  checkManifestAndWeightFiles(e) {
    let t = [], n = this.weightsFiles.map((r) => gx(r.name)), s = {};
    for (let r of e)
      r.paths.forEach((a) => {
        let i = gx(a);
        if (t.indexOf(i) !== -1)
          throw new Error(`Duplicate file basename found in weights manifest: '${i}'`);
        if (t.push(i), n.indexOf(i) === -1)
          throw new Error(`Weight file with basename '${i}' is not provided.`);
        s[a] = this.weightsFiles[n.indexOf(i)];
      });
    if (t.length !== this.weightsFiles.length)
      throw new Error(`Mismatch in the number of files in weights manifest (${t.length}) and the number of weight files provided (${this.weightsFiles.length}).`);
    return s;
  }
};
var sA = (e) => K().getBool("IS_BROWSER") && !Array.isArray(e) && e.startsWith(kd.URL_SCHEME) ? rA(e.slice(kd.URL_SCHEME.length)) : null;
xt.registerSaveRouter(sA);
function rA(e = "model") {
  return new kd(e);
}
function aA(e) {
  return new nA(e);
}
function yx(e, t, n, s) {
  i(e), n = n == null ? 0 : n, s = s == null ? 1 : s, o(n, s);
  let r = 0, a = (u) => (u.then((l) => {
    let c = n + ++r / e.length * (s - n);
    return t(c), l;
  }), u);
  function i(u) {
    F(u != null && Array.isArray(u) && u.length > 0, () => "promises must be a none empty array");
  }
  function o(u, l) {
    F(u >= 0 && u <= 1, () => `Progress fraction must be in range [0, 1], but got startFraction ${u}`), F(l >= 0 && l <= 1, () => `Progress fraction must be in range [0, 1], but got endFraction ${l}`), F(l >= u, () => `startFraction must be no more than endFraction, but got startFraction ${u} and endFraction ${l}`);
  }
  return Promise.all(e.map(a));
}
async function Rk(e, t) {
  t == null && (t = {});
  let n = t.fetchFunc == null ? K().platform.fetch : t.fetchFunc, s = e.map((p) => n(p, t.requestInit, { isBinary: true })), r = 0, a = 0.5, o = (t.onProgress == null ? await Promise.all(s) : await yx(s, t.onProgress, r, a)).map((p) => p.arrayBuffer()), u = 0.5, l = 1;
  return t.onProgress == null ? await Promise.all(o) : await yx(o, t.onProgress, u, l);
}
async function iA(e, t = "", n, s) {
  return Dk((i) => Rk(i, { requestInit: s }))(e, t, n);
}
function Dk(e) {
  return async (t, n = "", s) => {
    let r = t.map(() => false), a = {}, i = s != null ? s.map(() => false) : [], o = [];
    if (t.forEach((h, f) => {
      let m = 0;
      h.weights.forEach((g) => {
        let b = "quantization" in g ? g.quantization.dtype : g.dtype, y = lm[b] * dt(g.shape), v = () => {
          r[f] = true, a[f] == null && (a[f] = []), a[f].push({ manifestEntry: g, groupOffset: m, sizeBytes: y });
        };
        s != null ? s.forEach((x, k) => {
          x === g.name && (v(), i[k] = true);
        }) : v(), o.push(g.name), m += y;
      });
    }), !i.every((h) => h)) {
      let h = s.filter((f, m) => !i[m]);
      throw new Error(`Could not find weights in manifest with names: ${h.join(", ")}. 
Manifest JSON has weights with names: ${o.join(", ")}.`);
    }
    let u = r.reduce((h, f, m) => (f && h.push(m), h), []), l = [];
    u.forEach((h) => {
      t[h].paths.forEach((f) => {
        let m = n + (n.endsWith("/") ? "" : "/") + f;
        l.push(m);
      });
    });
    let c = await e(l), p = {}, d = 0;
    return u.forEach((h) => {
      let f = t[h].paths.length, m = 0;
      for (let x = 0; x < f; x++)
        m += c[d + x].byteLength;
      let g = new ArrayBuffer(m), b = new Uint8Array(g), y = 0;
      for (let x = 0; x < f; x++) {
        let k = new Uint8Array(c[d + x]);
        b.set(k, y), y += k.byteLength;
      }
      a[h].forEach((x) => {
        let k = g.slice(x.groupOffset, x.groupOffset + x.sizeBytes), I = Sk(k, [x.manifestEntry]);
        for (let $ in I)
          p[$] = I[$];
      }), d += f;
    }), p;
  };
}
var oA = "application/octet-stream";
var uA = "application/json";
var Gg = class {
  constructor(e, t) {
    if (this.DEFAULT_METHOD = "POST", t == null && (t = {}), this.weightPathPrefix = t.weightPathPrefix, this.onProgress = t.onProgress, this.weightUrlConverter = t.weightUrlConverter, t.fetchFunc != null ? (F(typeof t.fetchFunc == "function", () => "Must pass a function that matches the signature of `fetch` (see https://developer.mozilla.org/en-US/docs/Web/API/Fetch_API)"), this.fetch = t.fetchFunc) : this.fetch = K().platform.fetch, F(e != null && e.length > 0, () => "URL path for http must not be null, undefined or empty."), Array.isArray(e) && F(e.length === 2, () => `URL paths for http must have a length of 2, (actual length is ${e.length}).`), this.path = e, t.requestInit != null && t.requestInit.body != null)
      throw new Error("requestInit is expected to have no pre-existing body, but has one.");
    this.requestInit = t.requestInit || {};
  }
  async save(e) {
    if (e.modelTopology instanceof ArrayBuffer)
      throw new Error("BrowserHTTPRequest.save() does not support saving model topology in binary formats yet.");
    let t = Object.assign({ method: this.DEFAULT_METHOD }, this.requestInit);
    t.body = new FormData();
    let n = [{ paths: ["./model.weights.bin"], weights: e.weightSpecs }], s = Ik(e, n);
    t.body.append("model.json", new Blob([JSON.stringify(s)], { type: uA }), "model.json"), e.weightData != null && t.body.append("model.weights.bin", new Blob([e.weightData], { type: oA }), "model.weights.bin");
    let r = await this.fetch(this.path, t);
    if (r.ok)
      return { modelArtifactsInfo: zl(e), responses: [r] };
    throw new Error(`BrowserHTTPRequest.save() failed due to HTTP response status ${r.status}.`);
  }
  async load() {
    let e = await this.fetch(this.path, this.requestInit);
    if (!e.ok)
      throw new Error(`Request to ${this.path} failed with status code ${e.status}. Please verify this URL points to the model JSON of the model to load.`);
    let t;
    try {
      t = await e.json();
    } catch (r) {
      let a = `Failed to parse model JSON of response from ${this.path}.`;
      throw this.path.endsWith(".pb") ? a += " Your path contains a .pb file extension. Support for .pb models have been removed in TensorFlow.js 1.0 in favor of .json models. You can re-convert your Python TensorFlow model using the TensorFlow.js 1.0 conversion scripts or you can convert your.pb models with the 'pb2json'NPM script in the tensorflow/tfjs-converter repository." : a += " Please make sure the server is serving valid JSON for this request.", new Error(a);
    }
    let n = t.modelTopology, s = t.weightsManifest;
    if (n == null && s == null)
      throw new Error(`The JSON from HTTP path ${this.path} contains neither model topology or manifest for weights.`);
    return Ug(t, (r) => this.loadWeights(r));
  }
  async loadWeights(e) {
    let t = Array.isArray(this.path) ? this.path[1] : this.path, [n, s] = lA(t), r = this.weightPathPrefix || n, a = [];
    for (let l of e)
      a.push(...l.weights);
    let i = [], o = [];
    for (let l of e)
      for (let c of l.paths)
        this.weightUrlConverter != null ? o.push(this.weightUrlConverter(c)) : i.push(r + c + s);
    this.weightUrlConverter && i.push(...await Promise.all(o));
    let u = await Rk(i, { requestInit: this.requestInit, fetchFunc: this.fetch, onProgress: this.onProgress });
    return [a, Wg(u)];
  }
};
Gg.URL_SCHEME_REGEX = /^https?:\/\//;
function lA(e) {
  let t = e.lastIndexOf("/"), n = e.lastIndexOf("?"), s = e.substring(0, t), r = n > t ? e.substring(n) : "";
  return [s + "/", r];
}
function fm(e) {
  return e.match(Gg.URL_SCHEME_REGEX) != null;
}
var Fk = (e, t) => {
  if (typeof fetch == "undefined" && (t == null || t.fetchFunc == null))
    return null;
  {
    let n = true;
    if (Array.isArray(e) ? n = e.every((s) => fm(s)) : n = fm(e), n)
      return Hg(e, t);
  }
  return null;
};
xt.registerSaveRouter(Fk);
xt.registerLoadRouter(Fk);
function Hg(e, t) {
  return new Gg(e, t);
}
function cA(e, t) {
  return Hg(e, t);
}
var Gf = class {
  constructor(e) {
    this.modelArtifacts = e;
  }
  async load() {
    return this.modelArtifacts;
  }
};
var dA = class {
  constructor(e) {
    this.saveHandler = e;
  }
  async save(e) {
    return this.saveHandler(e);
  }
};
function pA(e, t, n, s) {
  return arguments.length === 1 ? e.modelTopology != null || e.weightSpecs != null ? new Gf(e) : (console.warn("Please call tf.io.fromMemory() with only one argument. The argument should be of type ModelArtifacts. The multi-argument signature of tf.io.fromMemory() has been deprecated and will be removed in a future release."), new Gf({ modelTopology: e })) : (console.warn("Please call tf.io.fromMemory() with only one argument. The argument should be of type ModelArtifacts. The multi-argument signature of tf.io.fromMemory() has been deprecated and will be removed in a future release."), new Gf({ modelTopology: e, weightSpecs: t, weightData: n, trainingConfig: s }));
}
function hA(e) {
  return new dA(e);
}
var fA = {};
Ee(fA, { confusionMatrix: () => vA });
function mA(e, t, n = false, s = false) {
  let r = _(e, "a", "matMul"), a = _(t, "b", "matMul");
  [r, a] = vt(r, a);
  let i = { a: r, b: a }, o = { transposeA: n, transposeB: s };
  return M.runKernel(Na, i, o);
}
var Ve = L({ matMul_: mA });
function gA(e, t, n = 1, s = 0) {
  if (t < 2)
    throw new Error(`Error in oneHot: depth must be >=2, but it is ${t}`);
  let a = { indices: _(e, "indices", "oneHot", "int32") }, i = { depth: t, onValue: n, offValue: s };
  return M.runKernel(Do, a, i);
}
var Sd = L({ oneHot_: gA });
function bA(e, t) {
  let n = _(e, "x", "transpose");
  if (t == null && (t = n.shape.map((a, i) => i).reverse()), F(n.rank === t.length, () => `Error in transpose: rank of input ${n.rank} must match length of perm ${t}.`), t.forEach((a) => {
    F(a >= 0 && a < n.rank, () => `All entries in 'perm' must be between 0 and ${n.rank - 1} but got ${t}`);
  }), n.rank <= 1)
    return n.clone();
  let s = { x: n }, r = { perm: t };
  return M.runKernel(mi, s, r);
}
var Ge = L({ transpose_: bA });
function yA(e, t, n) {
  let s = _(e, "labels", "confusionMatrix"), r = _(t, "predictions", "confusionMatrix");
  F(n == null || n > 0 && Number.isInteger(n), () => `If provided, numClasses must be a positive integer, but got ${n}`), F(s.rank === 1, () => `Expected the rank of labels to be 1, but got ${s.rank}`), F(r.rank === 1, () => `Expected the rank of predictions to be 1, but got ${r.rank}`), F(s.shape[0] === r.shape[0], () => `Mismatch in the number of examples: ${s.shape[0]} vs. ${r.shape[0]}. Labels and predictions should have the same number of elements.`), F(n > 0 && Number.isInteger(n), () => `numClasses is required to be a positive integer, but got ${n}`);
  let a = Sd(le(s, "int32"), n), i = Sd(le(r, "int32"), n), o = Ge(a), u = Ve(o, i);
  return le(u, "int32");
}
var vA = L({ confusionMatrix_: yA });
var Qo = {};
Ee(Qo, { assertAndGetBroadcastShape: () => rt, getBroadcastDims: () => Ok, getReductionAxes: () => At });
function Ok(e, t) {
  let n = e.length, s = [];
  for (let r = 0; r < n; r++) {
    let a = n - 1 - r, i = e[a] || 1;
    (t[t.length - 1 - r] || 1) > 1 && i === 1 && s.unshift(a);
  }
  return s;
}
function At(e, t) {
  let n = [];
  for (let s = 0; s < t.length; s++) {
    let r = e[e.length - s - 1], a = t.length - s - 1, i = t[a];
    (r == null || r === 1 && i > 1) && n.unshift(a);
  }
  return n;
}
function rt(e, t) {
  let n = [], s = Math.max(e.length, t.length);
  for (let r = 0; r < s; r++) {
    let a = e[e.length - r - 1];
    a == null && (a = 1);
    let i = t[t.length - r - 1];
    if (i == null && (i = 1), a === 1)
      n.unshift(i);
    else if (i === 1)
      n.unshift(a);
    else if (a !== i) {
      let o = `Operands could not be broadcast together with shapes ${e} and ${t}.`;
      throw Error(o);
    } else
      n.unshift(a);
  }
  return n;
}
var Pk = {};
Ee(Pk, { fromPixels: () => TA, fromPixelsAsync: () => CA, toPixels: () => NA });
function xA(e, t, n) {
  if (ka(e), t != null && t.length !== 3)
    throw new Error("tensor3d() requires shape to have three numbers");
  let s = Rs(e, n);
  if (s.length !== 3 && s.length !== 1)
    throw new Error("tensor3d() requires values to be number[][][] or flat/TypedArray");
  if (s.length === 1 && t == null)
    throw new Error("tensor3d() requires shape to be provided when `values` are a flat array");
  return Nr(e, t, s, n);
}
var Wr;
function zk(e, t = 3) {
  if (t > 4)
    throw new Error("Cannot construct Tensor with more than 4 channels from pixels.");
  if (e == null)
    throw new Error("pixels passed to tf.browser.fromPixels() can not be null");
  let n = false, s = false, r = false, a = false, i = false, o = false;
  if (e.data instanceof Uint8Array)
    n = true;
  else if (typeof ImageData != "undefined" && e instanceof ImageData)
    s = true;
  else if (typeof HTMLVideoElement != "undefined" && e instanceof HTMLVideoElement)
    r = true;
  else if (typeof HTMLImageElement != "undefined" && e instanceof HTMLImageElement)
    a = true;
  else if (e.getContext != null)
    i = true;
  else if (typeof ImageBitmap != "undefined" && e instanceof ImageBitmap)
    o = true;
  else
    throw new Error(`pixels passed to tf.browser.fromPixels() must be either an HTMLVideoElement, HTMLImageElement, HTMLCanvasElement, ImageData in browser, or OffscreenCanvas, ImageData in webworker or {data: Uint32Array, width: number, height: number}, but was ${e.constructor.name}`);
  if (r && r && e.readyState < 2)
    throw new Error("The video element has not loaded data yet. Please wait for `loadeddata` event on the <video> element.");
  if (rm(yd, M.backendName) != null) {
    let f = { pixels: e }, m = { numChannels: t };
    return M.runKernel(yd, f, m);
  }
  let [l, c] = r ? [e.videoWidth, e.videoHeight] : [e.width, e.height], p;
  if (i)
    p = e.getContext("2d").getImageData(0, 0, l, c).data;
  else if (s || n)
    p = e.data;
  else if (a || r || o) {
    if (Wr == null)
      if (typeof document == "undefined")
        if (typeof OffscreenCanvas != "undefined" && typeof OffscreenCanvasRenderingContext2D != "undefined")
          Wr = new OffscreenCanvas(1, 1).getContext("2d");
        else
          throw new Error("Cannot parse input in current context. Reason: OffscreenCanvas Context2D rendering is not supported.");
      else
        Wr = document.createElement("canvas").getContext("2d");
    Wr.canvas.width = l, Wr.canvas.height = c, Wr.drawImage(e, 0, 0, l, c), p = Wr.getImageData(0, 0, l, c).data;
  }
  let d;
  if (t === 4)
    d = new Int32Array(p);
  else {
    let f = l * c;
    d = new Int32Array(f * t);
    for (let m = 0; m < f; m++)
      for (let g = 0; g < t; ++g)
        d[m * t + g] = p[m * 4 + g];
  }
  return xA(d, [c, l, t], "int32");
}
function wA(e) {
  return e != null && e.data instanceof Uint8Array;
}
function kA() {
  return typeof window != "undefined" && typeof ImageBitmap != "undefined" && window.hasOwnProperty("createImageBitmap");
}
function SA(e) {
  return e != null && e.width !== 0 && e.height !== 0;
}
function IA(e) {
  return kA() && !(e instanceof ImageBitmap) && SA(e) && !wA(e);
}
async function CA(e, t = 3) {
  let n = null;
  if (K().getBool("WRAP_TO_IMAGEBITMAP") && IA(e)) {
    let s;
    try {
      s = await createImageBitmap(e, { premultiplyAlpha: "none" });
    } catch (r) {
      s = null;
    }
    s != null && s.width === e.width && s.height === e.height ? n = s : n = e;
  } else
    n = e;
  return zk(n, t);
}
async function NA(e, t) {
  let n = _(e, "img", "toPixels");
  if (!(e instanceof et)) {
    let l = n;
    n = le(l, "int32"), l.dispose();
  }
  if (n.rank !== 2 && n.rank !== 3)
    throw new Error(`toPixels only supports rank 2 or 3 tensors, got rank ${n.rank}.`);
  let [s, r] = n.shape.slice(0, 2), a = n.rank === 2 ? 1 : n.shape[2];
  if (a > 4 || a === 2)
    throw new Error(`toPixels only supports depth of size 1, 3 or 4 but got ${a}`);
  if (n.dtype !== "float32" && n.dtype !== "int32")
    throw new Error(`Unsupported type for toPixels: ${n.dtype}. Please use float32 or int32 tensors.`);
  let i = await n.data(), o = n.dtype === "float32" ? 255 : 1, u = new Uint8ClampedArray(r * s * 4);
  for (let l = 0; l < s * r; ++l) {
    let c = [0, 0, 0, 255];
    for (let d = 0; d < a; d++) {
      let h = i[l * a + d];
      if (n.dtype === "float32") {
        if (h < 0 || h > 1)
          throw new Error(`Tensor values for a float32 Tensor must be in the range [0 - 1] but encountered ${h}.`);
      } else if (n.dtype === "int32" && (h < 0 || h > 255))
        throw new Error(`Tensor values for a int32 Tensor must be in the range [0 - 255] but encountered ${h}.`);
      a === 1 ? (c[0] = h * o, c[1] = h * o, c[2] = h * o) : c[d] = h * o;
    }
    let p = l * 4;
    u[p + 0] = Math.round(c[0]), u[p + 1] = Math.round(c[1]), u[p + 2] = Math.round(c[2]), u[p + 3] = Math.round(c[3]);
  }
  if (t != null) {
    t.width = r, t.height = s;
    let l = t.getContext("2d"), c = new ImageData(u, r, s);
    l.putImageData(c, 0, 0);
  }
  return n !== e && n.dispose(), u;
}
var TA = L({ fromPixels_: zk });
var Mk = {};
Ee(Mk, { prepareAndValidate: () => Lk });
function Lk(e, t) {
  let n = e.shape.length, s = t.shape.length;
  if (n < 1)
    throw new Error(`tf.gatherND() expects the input to be rank 1 or higher, but the rank was ${n}.`);
  if (s < 1)
    throw new Error(`tf.gatherND() expects the indices to be rank 1 or higher, but the rank was ${s}.`);
  if (t.dtype !== "int32")
    throw new Error(`tf.gatherND() expects the indices to be int32 type, but the dtype was ${t.dtype}.`);
  if (t.shape[s - 1] > n)
    throw new Error(`index innermost dimension length must be <= tensor rank; saw: ${t.shape[s - 1]} vs. ${n}`);
  if (dt(e.shape) === 0)
    throw new Error(`Requested more than 0 entries, but input is empty. Input shape: ${e.shape}.`);
  let r = t.shape, a = r[r.length - 1], i = 1;
  for (let p = 0; p < r.length - 1; ++p)
    i *= r[p];
  let o = e.shape, u = r.slice();
  u.pop();
  let l = 1;
  for (let p = a; p < n; ++p)
    l *= o[p], u.push(o[p]);
  let c = [...co(e.shape).map((p) => p / l), 1].slice(0, a);
  return [u, i, l, c];
}
var Bk = {};
Ee(Bk, { calculateShapes: () => Vk, validateInput: () => jg, validateUpdateShape: () => qg });
function qg(e, t, n) {
  let s = t.rank > 1 ? t.shape[t.rank - 1] : 1, r = t.rank > 1 ? t.rank - 1 : 1, a = `Must have updates.shape = indices.shape[:batchDim] + shape[sliceDim:], got updates.shape: ${n.shape}, indices.shape: ${t.shape}, shape: ${e}, sliceDim: ${s}, and batchDim: ${r}.`;
  if (n.rank < r)
    throw new Error(a + ` update.rank < ${r}. `);
  if (e.length < s + (n.rank - r))
    throw new Error(a + ` Output shape length < ${s + (n.rank - r)}`);
  if (n.rank !== r + e.length - s)
    throw new Error(a + ` update.rank != ${r + e.length - s}`);
  for (let i = 0; i < r; ++i)
    if (n.shape[i] !== t.shape[i])
      throw new Error(a + ` updates.shape[${i}] (${n.shape[i]}) != indices.shape[${i}] (${t.shape[i]}).`);
  for (let i = 0; i < n.rank - r; ++i)
    if (n.shape[i + r] !== e[i + s])
      throw new Error(a + ` updates.shape[${i + r}] (${n.shape[i + r]}) != shape[${i + r}] (${e[i + r]})`);
}
function jg(e, t, n) {
  if (t.rank < 1)
    throw new Error(`tf.scatterND() expects the indices to be rank 1 or higher, but the rank was ${t.rank}.`);
  if (e.rank < 1)
    throw new Error(`tf.scatterND() expects the updates to be rank 1 or higher, but the rank was ${e.rank}.`);
  if (t.dtype !== "int32")
    throw new Error(`The dtype of 'indices' should be int32, but got dtype: ${t.dtype}`);
  if (n.length < 1)
    throw new Error(`Output rank must be greater or equal to 1, but got shape: ${n}`);
  if (n.length === 0) {
    if (t.size === 0)
      throw new Error(`Indices specified for empty output. indices shape: ${t.shape}`);
    if (e.size === 0)
      throw new Error(`Updates specified for empty output. updates shape: ${e.shape}`);
  }
  qg(n, t, e);
}
function Vk(e, t, n) {
  let s = t.shape.length, r = s > 1 ? t.shape[s - 1] : 1, a = n.length, i = 1;
  for (let p = r; p < a; ++p)
    i *= n[p];
  let o = r < 1 ? 1 : r, u = dt(t.shape) / o, l = [...co(n.slice(0, r)), 1], c = dt(n);
  return { sliceRank: r, numUpdates: u, sliceSize: i, strides: l, outputSize: c };
}
var wt = {};
Ee(wt, { assertParamsValid: () => _A, computeFlatOffset: () => FA, computeOutShape: () => EA, getNormalizedAxes: () => RA, isSliceContinous: () => DA, maskToAxes: () => AA, parseSliceParams: () => Yk, sliceInfo: () => OA, startForAxis: () => Kk, startIndicesWithElidedDims: () => Hk, stopForAxis: () => Xk, stopIndicesWithElidedDims: () => qk, stridesForAxis: () => jk, stridesWithElidedDims: () => Wk });
var mm = -2;
var $A = -1;
function _A(e, t, n) {
  let s = e.shape.length;
  F(s === t.length, () => `Error in slice${s}D: Length of begin ${t} must match the rank of the array (${s}).`), F(s === n.length, () => `Error in slice${s}D: Length of size ${n} must match the rank of the array (${s}).`);
  for (let r = 0; r < s; ++r)
    F(t[r] + n[r] <= e.shape[r], () => `Error in slice${s}D: begin[${r}] + size[${r}] (${t[r] + n[r]}) would overflow input.shape[${r}] (${e.shape[r]})`);
}
function AA(e) {
  let t = [], n = 0;
  for (; e > 0; )
    e & 1 && t.push(n), e /= 2, n++;
  return t;
}
function EA(e, t, n) {
  let s = [];
  for (let r = 0; r < e.length; r++)
    s[r] = Math.ceil((t[r] - e[r]) / n[r]);
  return s;
}
function Wk(e, t, n, s) {
  let r = [...e];
  for (let a = r.length; a < s.length; a++)
    r.push(1);
  for (let a = 0; a < n; a++)
    a === 0 ? r[t] = 1 : (r.splice(t, 0, 1), r.pop());
  return r;
}
function Uk(e, t, n) {
  return n <= e ? n : n - (t - 1);
}
function Gk(e, t) {
  let n = [];
  for (let s = 0; s < e; s++)
    n.push(t + s);
  return n;
}
function RA(e, t, n, s, r, a, i, o, u) {
  let l = e.length, c = new Array(l), p = new Array(l), d = new Array(l);
  if (t.length && n > 0) {
    let h = t[0], f = n + 1;
    c = Hk(i, h, f, s, e), p = qk(o, h, f, r, e), d = Wk(a, h, f, e);
  } else
    for (let h = 0; h < l; h++)
      c[h] = Kk(i, s, a, e, h, u), p[h] = Xk(o, r, a, e, h, u), d[h] = jk(a, h, u);
  return { begin: c, end: p, strides: d };
}
function Hk(e, t, n, s, r) {
  let a = [...r], i = Gk(n, t);
  for (let o = 0; o < a.length; o++)
    if (i.indexOf(o) > -1)
      a[o] = 0;
    else {
      let u = Uk(t, n, o), l = s[u];
      e & 1 << u && (l = 0), a[o] = l;
    }
  return a;
}
function qk(e, t, n, s, r) {
  let a = [...r], i = Gk(n, t);
  for (let o = 0; o < a.length; o++)
    if (i.indexOf(o) > -1)
      a[o] = Number.MAX_SAFE_INTEGER;
    else {
      let u = Uk(t, n, o), l = s[u];
      e & 1 << u && (l = Number.MAX_SAFE_INTEGER), a[o] = l;
    }
  for (let o = 0; o < a.length; o++) {
    let u = r[o];
    a[o] < 0 && (a[o] += u), a[o] = Hu(0, a[o], r[o]);
  }
  return a;
}
function jk(e, t, n) {
  let s = e[t];
  return (n & 1 << t || s == null) && (s = 1), s;
}
function Kk(e, t, n, s, r, a) {
  let i = t[r], o = n[r] || 1;
  (e & 1 << r || a & 1 << r || i == null) && (o > 0 ? i = Number.MIN_SAFE_INTEGER : i = Number.MAX_SAFE_INTEGER);
  let u = s[r];
  return i < 0 && (i += u), i = Hu(0, i, u - 1), i;
}
function Xk(e, t, n, s, r, a) {
  let i = t[r], o = n[r] || 1;
  (e & 1 << r || a & 1 << r || i == null) && (o > 0 ? i = Number.MAX_SAFE_INTEGER : i = Number.MIN_SAFE_INTEGER);
  let u = s[r];
  return i < 0 && (i += u), o > 0 ? i = Hu(0, i, u) : i = Hu(-1, i, u - 1), i;
}
function DA(e, t, n) {
  let s = n.length;
  for (let r = 0; r < n.length; r++)
    if (n[r] > 1) {
      s = r;
      break;
    }
  for (let r = s + 1; r < n.length; r++)
    if (t[r] > 0 || n[r] !== e[r])
      return false;
  return true;
}
function FA(e, t) {
  let n = e.length > 0 ? e[e.length - 1] : 1;
  for (let s = 0; s < e.length - 1; s++)
    n += e[s] * t[s];
  return n;
}
function Yk(e, t, n) {
  let s, r = e.shape.length;
  typeof t == "number" ? s = [t, ...new Array(r - 1).fill(0)] : t.length < r ? s = t.concat(new Array(r - t.length).fill(0)) : s = t.slice(), s.forEach((i) => {
    F(i !== -1, () => "slice() does not support negative begin indexing.");
  });
  let a;
  return n == null ? a = new Array(r).fill(-1) : typeof n == "number" ? a = [n, ...new Array(r - 1).fill(-1)] : n.length < r ? a = n.concat(new Array(r - n.length).fill(-1)) : a = n, a = a.map((i, o) => i >= 0 ? i : (F(i === -1, () => `Negative size values should be exactly -1 but got ${i} for the slice() size at index ${o}.`), e.shape[o] - s[o])), [s, a];
}
function OA(e, t, n, s, r, a, i, o, u) {
  let l;
  if (s == null ? (l = new Array(t.length), l.fill(1)) : l = s, i != null && (i & i - 1) !== 0)
    throw new Error("Multiple ellipses in slice is not allowed.");
  let c = false, p = { dims: l.length, numAddAxisAfterEllipsis: 0, begin: t.slice(), end: n.slice(), strides: l.slice(), beginMask: r, endMask: a, ellipsisMask: i, newAxisMask: o, shrinkAxisMask: u };
  for (let v = 0; v < p.dims; v++)
    c && (1 << v & o) !== 0 && p.numAddAxisAfterEllipsis++, 1 << v & i && (c = true);
  c || (p.ellipsisMask |= 1 << p.dims, p.dims++);
  let d = { dims: e.length, beginMask: 0, endMask: 0, beginValid: false, endValid: false };
  PA(p, d);
  let h = true, f = true, m = true, g = [], b = [];
  for (let v = 0; v < e.length; ++v) {
    if (d.strides[v] === 0)
      throw Error(`strides[${v}] must be non-zero`);
    let x = !!(d.shrinkAxisMask & 1 << v), k = e[v];
    if (k === -1) {
      g.push(x ? 1 : -1);
      continue;
    }
    let I = [d.beginMask & 1 << v, d.endMask & 1 << v], $ = [d.strides[v] > 0 ? 0 : -1, d.strides[v] > 0 ? k : k - 1];
    if (x && d.strides[v] <= 0)
      throw Error("only stride 1 allowed on non-range indexing.");
    m = m && d.strides[v] === 1;
    let R = !!(d.beginMask & 1 << v && d.endMask & 1 << v);
    if (d.beginValid && d.endValid) {
      if (x) {
        let O = d.begin[v] < 0 ? k + d.begin[v] : d.begin[v];
        if (d.begin[v] = O, d.end[v] = d.begin[v] + 1, O < 0 || O >= k)
          throw Error(`slice index ${d.begin[v]} of dimension ${v} out of bounds.`);
      } else
        d.begin[v] = vx(d.begin[v], 0, d.strides[v], k, I, $), d.end[v] = vx(d.end[v], 1, d.strides[v], k, I, $);
      let A = d.strides[v] === 1 && d.begin[v] === 0 && d.end[v] === k;
      h = h && A, f = f && (v === 0 && d.strides[v] === 1 || A);
    } else
      h = h && d.strides[v] === 1 && R, f = f && (v === 0 && d.strides[v] === 1 || R);
    let E, P = false;
    if (d.beginValid && d.endValid ? (E = d.end[v] - d.begin[v], P = true) : x ? (E = 1, P = true) : R && k >= 0 && (d.strides[v] < 0 ? E = -k : E = k, P = true), P) {
      let A;
      E === 0 || E < 0 != d.strides[v] < 0 ? A = 0 : A = Math.trunc(E / d.strides[v]) + (E % d.strides[v] !== 0 ? 1 : 0), g.push(A);
    } else
      g.push(-1);
  }
  for (let v = 0; v < d.finalShapeGatherIndices.length; ++v) {
    let x = d.finalShapeGatherIndices[v];
    x >= 0 ? b.push(g[x]) : x === mm && b.push(1);
  }
  return { finalShapeSparse: b.filter((v, x) => d.finalShapeGatherIndices[x] !== mm), finalShape: b, isIdentity: h, sliceDim0: f, isSimpleSlice: m, begin: d.begin, end: d.end, strides: d.strides };
}
function PA(e, t) {
  t.beginMask = 0, t.endMask = 0, t.shrinkAxisMask = 0;
  let n = 0;
  t.beginValid = e.begin != null, t.endValid = e.end != null, t.begin = new Array(t.dims), t.end = new Array(t.dims), t.strides = new Array(t.dims), t.finalShapeGatherIndices = [], t.finalShapeGatherIndicesSparse = [], t.inputShapeGatherIndicesSparse = new Array(t.dims);
  for (let s = 0; s < e.dims; s++)
    if (1 << s & e.ellipsisMask) {
      let r = Math.min(t.dims - (e.dims - s) + 1 + e.numAddAxisAfterEllipsis, t.dims);
      for (; n < r; n++)
        t.begin[n] = 0, t.end[n] = 0, t.strides[n] = 1, t.beginMask |= 1 << n, t.endMask |= 1 << n, t.finalShapeGatherIndices.push(n), t.finalShapeGatherIndicesSparse.push(-1), t.inputShapeGatherIndicesSparse[n] = s;
    } else if (1 << s & e.newAxisMask)
      t.finalShapeGatherIndices.push(mm), t.finalShapeGatherIndicesSparse.push(-1);
    else {
      if (n === t.begin.length)
        throw Error(`Index out of range using input dim ${n}; input has only ${t.dims} dims, ${t.begin.length}.`);
      e.begin != null && (t.begin[n] = e.begin[s]), e.end != null && (t.end[n] = e.end[s]), t.strides[n] = e.strides[s], e.beginMask & 1 << s && (t.beginMask |= 1 << n), e.endMask & 1 << s && (t.endMask |= 1 << n), e.shrinkAxisMask & 1 << s ? (t.finalShapeGatherIndices.push($A), t.finalShapeGatherIndicesSparse.push(-1), t.shrinkAxisMask |= 1 << n) : (t.finalShapeGatherIndices.push(n), t.finalShapeGatherIndicesSparse.push(s)), t.inputShapeGatherIndicesSparse[n] = s, n++;
    }
}
function vx(e, t, n, s, r, a) {
  if (r[t])
    return n > 0 ? a[t] : a[t + 1 & 1];
  {
    let i = e < 0 ? s + e : e;
    return i < a[0] ? a[0] : i > a[1] ? a[1] : i;
  }
}
var re = {};
Ee(re, { Serializable: () => Qk, SerializationMap: () => Kr, registerClass: () => Tr });
var Qk = class {
  getClassName() {
    return this.constructor.className;
  }
  static fromConfig(e, t) {
    return new e(t);
  }
};
var Kr = class {
  constructor() {
    this.classNameMap = {};
  }
  static getMap() {
    return Kr.instance == null && (Kr.instance = new Kr()), Kr.instance;
  }
  static register(e) {
    Kr.getMap().classNameMap[e.className] = [e, e.fromConfig];
  }
};
function Tr(e) {
  F(e.className != null, () => "Class being registered does not have the static className property defined."), F(typeof e.className == "string", () => "className is required to be a string, but got type " + typeof e.className), F(e.className.length > 0, () => "Class being registered has an empty-string as its className, which is disallowed."), Kr.register(e);
}
var zA = {};
Ee(zA, { TEST_EPSILON_FLOAT16: () => Zk, encodeStrings: () => Jk, expectArrayBuffersEqual: () => GA, expectArraysClose: () => LA, expectArraysEqual: () => VA, expectNumbersClose: () => WA, expectPromiseToFail: () => BA, expectValuesInRange: () => UA, testEpsilon: () => Kg });
var MA = 1e-3;
var Zk = 0.1;
function LA(e, t, n) {
  return n == null && (n = Kg()), gm(e, t, (s, r) => Xg(s, r, n));
}
function Kg() {
  return M.backend.floatPrecision() === 32 ? MA : Zk;
}
function gm(e, t, n) {
  let s = true;
  if ((Qt(e) || Qt(t)) && (s = false), Qt(e) && Qt(t) && (s = true), s) {
    let i = e.constructor.name, o = t.constructor.name;
    if (i !== o)
      throw new Error(`Arrays are of different type. Actual: ${i}. Expected: ${o}`);
  }
  if (Array.isArray(e) && Array.isArray(t)) {
    let i = Rs(e), o = Rs(t);
    if (!kr(i, o))
      throw new Error(`Arrays have different shapes. Actual: [${i}]. Expected: [${o}]`);
  }
  let r = Qt(e) ? e : ra(e), a = Qt(t) ? t : ra(t);
  if (r.length !== a.length)
    throw new Error(`Arrays have different lengths actual: ${r.length} vs expected: ${a.length}.
Actual:   ${r}.
Expected: ${a}.`);
  for (let i = 0; i < a.length; ++i) {
    let o = r[i], u = a[i];
    if (!n(o, u))
      throw new Error(`Arrays differ: actual[${i}] = ${o}, expected[${i}] = ${u}.
Actual:   ${r}.
Expected: ${a}.`);
  }
}
function BA(e, t) {
  e().then(() => t.fail(), () => t());
}
function VA(e, t) {
  let n = typeof t == "string" || typeof t == "number" || typeof t == "boolean" ? [t] : t;
  return ar(e) || ar(e[0]) || ar(t) || ar(t[0]) ? gm(e, n, (s, r) => s == r) : gm(e, t, (s, r) => Xg(s, r, 0));
}
function WA(e, t, n) {
  if (n == null && (n = Kg()), !Xg(e, t, n))
    throw new Error(`Numbers differ: actual === ${e}, expected === ${t}`);
}
function Xg(e, t, n) {
  return !isFinite(e) && !isFinite(t) ? true : !(isNaN(e) || isNaN(t) || Math.abs(e - t) > n);
}
function UA(e, t, n) {
  for (let s = 0; s < e.length; s++)
    if (e[s] < t || e[s] > n)
      throw new Error(`Value out of range:${e[s]} low: ${t}, high: ${n}`);
}
function GA(e, t) {
  let n = new Float32Array(e), s = new Float32Array(t);
  if (n.length !== s.length)
    throw new Error(`Expected ArrayBuffer to be of length ${s.length}, but it was ${n.length}`);
  for (let r = 0; r < s.length; r++)
    if (n[r] !== s[r])
      throw new Error(`Expected ArrayBuffer value at ${r} to be ${s[r]} but got ${n[r]} instead`);
}
function Jk(e) {
  for (let t = 0; t < e.length; t++) {
    let n = e[t];
    Array.isArray(n) ? Jk(n) : e[t] = Pl(n);
  }
  return e;
}
var cpe = "0.0.0";
function dpe() {
  K().set("PROD", true);
}
function ppe() {
  K().set("DEBUG", true);
}
function hpe() {
  K().set("DEPRECATION_WARNINGS_ENABLED", false), console.warn("TensorFlow.js deprecation warnings have been disabled.");
}
function eS(e) {
  K().getBool("DEPRECATION_WARNINGS_ENABLED") && console.warn(e + " You can disable deprecation warnings with tf.disableDeprecationWarnings().");
}
i_(eS);
function fpe() {
  M.disposeVariables();
}
function ds() {
  return M;
}
function bm() {
  return M.memory();
}
function mpe(e) {
  return M.profile(e);
}
function j(e, t) {
  return M.tidy(e, t);
}
function De(e) {
  Lg(e).forEach((n) => n.dispose());
}
function qt(e) {
  return M.keep(e);
}
function gpe(e) {
  return M.time(e);
}
function bpe(e) {
  return M.setBackend(e);
}
function ype() {
  return M.ready();
}
function vpe() {
  return M.backendName;
}
function xpe(e) {
  M.removeBackend(e);
}
function wpe(e) {
  return M.findBackend(e);
}
function kpe(e) {
  return M.findBackendFactory(e);
}
function vp(e, t, n = 1) {
  return M.registerBackend(e, t, n);
}
function HA() {
  return M.backend;
}
function Spe(e, t) {
  K().setPlatform(e, t);
}
function qA(e, t) {
  let n = _(e, "a", "add"), s = _(t, "b", "add");
  [n, s] = vt(n, s);
  let r = { a: n, b: s };
  return M.runKernel(Sr, r);
}
var ie = L({ add_: qA });
function jA(e, t) {
  let n = _(e, "a", "floorDiv"), s = _(t, "b", "floorDiv");
  [n, s] = vt(n, s);
  let r = { a: n, b: s };
  return M.runKernel(La, r);
}
var tS = L({ floorDiv_: jA });
function KA(e, t) {
  let n = _(e, "a", "div"), s = _(t, "b", "div");
  if ([n, s] = vt(n, s), n.dtype === "int32" && s.dtype === "int32")
    return tS(n, s);
  let r = { a: n, b: s }, a = {};
  return M.runKernel(Oa, r, a);
}
var xe = L({ div_: KA });
function XA(e, t) {
  let n = _(e, "a", "mul"), s = _(t, "b", "mul");
  [n, s] = vt(n, s);
  let r = { a: n, b: s };
  return M.runKernel(Za, r);
}
var V = L({ mul_: XA });
function YA(e) {
  let t = _(e, "x", "abs");
  if (t.dtype === "complex64") {
    let n = { x: t };
    return M.runKernel(tp, n);
  } else {
    let n = { x: t };
    return M.runKernel(po, n);
  }
}
var Lt = L({ abs_: YA });
function QA(e) {
  let n = { x: _(e, "x", "acos") };
  return M.runKernel(ol, n);
}
var ZA = L({ acos_: QA });
function JA(e) {
  let n = { x: _(e, "x", "acosh") };
  return M.runKernel(ul, n);
}
var eE = L({ acosh_: JA });
function tE(e) {
  F(Array.isArray(e), () => "The argument passed to tf.addN() must be a list of tensors"), F(e.length >= 1, () => `Must pass at least one tensor to tf.addN(), but got ${e.length}`);
  let t = e.map((r, a) => _(r, `tensors${a}`, "addN")), n = t[0];
  t.forEach((r) => {
    if (r.dtype !== n.dtype)
      throw new Error("All tensors passed to tf.addN() must have the same dtype");
  }), t.forEach((r) => {
    if (!kr(r.shape, n.shape))
      throw new Error("All tensors passed to tf.addN() must have the same shape");
  });
  let s = t;
  return M.runKernel(Sa, s);
}
var nE = L({ addN_: tE });
function sE(e, t = null, n = false) {
  let r = { x: _(e, "x", "all", "bool") }, a = { axis: t, keepDims: n };
  return M.runKernel(ll, r, a);
}
var nS = L({ all_: sE });
function rE(e, t = null, n = false) {
  let r = { x: _(e, "x", "any", "bool") }, a = { axis: t, keepDims: n };
  return M.runKernel(cl, r, a);
}
var ym = L({ any_: rE });
function aE(e, t = 0) {
  let s = { x: _(e, "x", "argMax") }, r = { axis: t };
  return M.runKernel(Ia, s, r);
}
var Xu = L({ argMax_: aE });
function iE(e, t = 0) {
  let s = { x: _(e, "x", "argMin") }, r = { axis: t };
  return M.runKernel(dl, s, r);
}
var oE = L({ argMin_: iE });
function uE(e) {
  let n = { x: _(e, "x", "asin") };
  return M.runKernel(pl, n);
}
var lE = L({ asin_: uE });
function cE(e) {
  let n = { x: _(e, "x", "asinh") };
  return M.runKernel(hl, n);
}
var dE = L({ asinh_: cE });
function pE(e) {
  let n = { x: _(e, "x", "atan") };
  return M.runKernel(fl, n);
}
var hE = L({ atan_: pE });
function fE(e, t) {
  let n = _(e, "a", "atan2"), s = _(t, "b", "atan2");
  [n, s] = vt(n, s);
  let r = { a: n, b: s };
  return M.runKernel(gl, r);
}
var mE = L({ atan2_: fE });
function gE(e) {
  let n = { x: _(e, "x", "atanh") };
  return M.runKernel(ml, n);
}
var bE = L({ atanh_: gE });
function yE(e, t, n, s, r = "NHWC", a) {
  let i = e[3], o = [...t, i], u = aS(r);
  return Ml(e, o, n, a, s, null, null, u);
}
function sS(e, t, n, s, r, a, i = "channelsLast") {
  let [o, u] = Id(t), l;
  if (i === "channelsLast")
    l = [o, u, e[3], e[3]];
  else if (i === "channelsFirst")
    l = [o, u, e[1], e[1]];
  else
    throw new Error(`Unknown dataFormat ${i}`);
  return Ml(e, l, n, s, r, a, false, i);
}
function vE(e, t, n, s, r, a, i = "NDHWC") {
  let [o, u, l] = vm(t), c, p;
  if (i === "NDHWC")
    p = "channelsLast", c = [o, u, l, e[4], e[4]];
  else if (i === "NCDHW")
    p = "channelsFirst", c = [o, u, l, e[1], e[1]];
  else
    throw new Error(`Unknown dataFormat ${i}`);
  return rS(e, c, n, s, r, false, p, a);
}
function Ml(e, t, n, s, r, a, i = false, o = "channelsLast") {
  let [u, l, c, p] = [-1, -1, -1, -1];
  if (o === "channelsLast")
    [u, l, c, p] = e;
  else if (o === "channelsFirst")
    [u, p, l, c] = e;
  else
    throw new Error(`Unknown dataFormat ${o}`);
  let [d, h, , f] = t, [m, g] = Id(n), [b, y] = Id(s), v = Qi(d, b), x = Qi(h, y), { padInfo: k, outHeight: I, outWidth: $ } = kE(r, l, c, m, g, v, x, a, o), R = i ? f * p : f, E;
  return o === "channelsFirst" ? E = [u, R, I, $] : o === "channelsLast" && (E = [u, I, $, R]), { batchSize: u, dataFormat: o, inHeight: l, inWidth: c, inChannels: p, outHeight: I, outWidth: $, outChannels: R, padInfo: k, strideHeight: m, strideWidth: g, filterHeight: d, filterWidth: h, effectiveFilterHeight: v, effectiveFilterWidth: x, dilationHeight: b, dilationWidth: y, inShape: e, outShape: E, filterShape: t };
}
function rS(e, t, n, s, r, a = false, i = "channelsLast", o) {
  let [u, l, c, p, d] = [-1, -1, -1, -1, -1];
  if (i === "channelsLast")
    [u, l, c, p, d] = e;
  else if (i === "channelsFirst")
    [u, d, l, c, p] = e;
  else
    throw new Error(`Unknown dataFormat ${i}`);
  let [h, f, m, , g] = t, [b, y, v] = vm(n), [x, k, I] = vm(s), $ = Qi(h, x), R = Qi(f, k), E = Qi(m, I), { padInfo: P, outDepth: A, outHeight: O, outWidth: T } = SE(r, l, c, p, b, y, v, $, R, E, o), z = a ? g * d : g, W;
  return i === "channelsFirst" ? W = [u, z, A, O, T] : i === "channelsLast" && (W = [u, A, O, T, z]), { batchSize: u, dataFormat: i, inDepth: l, inHeight: c, inWidth: p, inChannels: d, outDepth: A, outHeight: O, outWidth: T, outChannels: z, padInfo: P, strideDepth: b, strideHeight: y, strideWidth: v, filterDepth: h, filterHeight: f, filterWidth: m, effectiveFilterDepth: $, effectiveFilterHeight: R, effectiveFilterWidth: E, dilationDepth: x, dilationHeight: k, dilationWidth: I, inShape: e, outShape: W, filterShape: t };
}
function xE(e, t, n, s, r) {
  s == null && (s = Yg(e, t, n));
  let a = e[0], i = e[1], o = ea((a - t + 2 * s) / n + 1, r), u = ea((i - t + 2 * s) / n + 1, r);
  return [o, u];
}
function wE(e, t, n, s, r, a) {
  r == null && (r = Yg(e, t, s));
  let i = e[0], o = e[1], u = e[2], l = ea((i - t + 2 * r) / s + 1, a), c = ea((o - t + 2 * r) / s + 1, a), p = ea((u - t + 2 * r) / s + 1, a);
  return [l, c, p, n];
}
function Yg(e, t, n, s = 1) {
  let r = Qi(t, s);
  return Math.floor((e[0] * (n - 1) - n + r) / 2);
}
function Id(e) {
  return typeof e == "number" ? [e, e, e] : e.length === 2 ? [e[0], e[1], 1] : e;
}
function vm(e) {
  return typeof e == "number" ? [e, e, e] : e;
}
function Qi(e, t) {
  return t <= 1 ? e : e + (e - 1) * (t - 1);
}
function kE(e, t, n, s, r, a, i, o, u) {
  let l, c, p;
  if (typeof e == "number") {
    l = { top: e, bottom: e, left: e, right: e, type: e === 0 ? "VALID" : "NUMBER" };
    let h = xE([t, n], a, s, e, o);
    c = h[0], p = h[1];
  } else if (e === "same") {
    c = Math.ceil(t / s), p = Math.ceil(n / r);
    let d = Math.max(0, (c - 1) * s + a - t), h = Math.max(0, (p - 1) * r + i - n), f = Math.floor(d / 2), m = d - f, g = Math.floor(h / 2), b = h - g;
    l = { top: f, bottom: m, left: g, right: b, type: "SAME" };
  } else if (e === "valid")
    l = { top: 0, bottom: 0, left: 0, right: 0, type: "VALID" }, c = Math.ceil((t - a + 1) / s), p = Math.ceil((n - i + 1) / r);
  else if (typeof e == "object") {
    let d = u === "channelsLast" ? e[1][0] : e[2][0], h = u === "channelsLast" ? e[1][1] : e[2][1], f = u === "channelsLast" ? e[2][0] : e[3][0], m = u === "channelsLast" ? e[2][1] : e[3][1];
    l = { top: d, bottom: h, left: f, right: m, type: d === 0 && h === 0 && f === 0 && m === 0 ? "VALID" : "EXPLICIT" }, c = ea((t - a + d + h) / s + 1, o), p = ea((n - i + f + m) / r + 1, o);
  } else
    throw Error(`Unknown padding parameter: ${e}`);
  return { padInfo: l, outHeight: c, outWidth: p };
}
function SE(e, t, n, s, r, a, i, o, u, l, c) {
  let p, d, h, f;
  if (typeof e == "number") {
    p = { top: e, bottom: e, left: e, right: e, front: e, back: e, type: e === 0 ? "VALID" : "NUMBER" };
    let g = wE([t, n, s, 1], o, 1, r, e, c);
    d = g[0], h = g[1], f = g[2];
  } else if (e === "same") {
    d = Math.ceil(t / r), h = Math.ceil(n / a), f = Math.ceil(s / i);
    let m = (d - 1) * r + o - t, g = (h - 1) * a + u - n, b = (f - 1) * i + l - s, y = Math.floor(m / 2), v = m - y, x = Math.floor(g / 2), k = g - x, I = Math.floor(b / 2), $ = b - I;
    p = { top: x, bottom: k, left: I, right: $, front: y, back: v, type: "SAME" };
  } else if (e === "valid")
    p = { top: 0, bottom: 0, left: 0, right: 0, front: 0, back: 0, type: "VALID" }, d = Math.ceil((t - o + 1) / r), h = Math.ceil((n - u + 1) / a), f = Math.ceil((s - l + 1) / i);
  else
    throw Error(`Unknown padding parameter: ${e}`);
  return { padInfo: p, outDepth: d, outHeight: h, outWidth: f };
}
function ea(e, t) {
  if (!t)
    return Math.trunc(e);
  switch (t) {
    case "round":
      return Math.round(e);
    case "ceil":
      return Math.ceil(e);
    case "floor":
      return Math.floor(e);
    default:
      throw new Error(`Unknown roundingMode ${t}`);
  }
}
function fr(e) {
  let [t, n, s] = Id(e);
  return t === 1 && n === 1 && s === 1;
}
function Ps(e, t) {
  return fr(e) || fr(t);
}
function aS(e) {
  if (e === "NHWC")
    return "channelsLast";
  if (e === "NCHW")
    return "channelsFirst";
  throw new Error(`Unknown dataFormat ${e}`);
}
function hn(e, t, n) {
  if (n != null) {
    if (typeof t == "string")
      throw Error(`Error in ${e}: pad must be an integer when using dimRoundingMode ${n} but got pad ${t}.`);
    if (typeof t == "number")
      F(eo(t), () => `Error in ${e}: pad must be an integer when using dimRoundingMode ${n} but got pad ${t}.`);
    else if (typeof t == "object")
      t.forEach((s) => {
        s.forEach((r) => {
          F(eo(r), () => `Error in ${e}: pad must be an integer when using dimRoundingMode ${n} but got pad ${r}.`);
        });
      });
    else
      throw Error(`Error in ${e}: Unknown padding parameter: ${t}`);
  }
}
function IE(e, t) {
  let s = { x: _(e, "x", "reshape", "string_or_numeric") }, r = { shape: t };
  return M.runKernel(Oo, s, r);
}
var U = L({ reshape_: IE });
function CE(e, t, n, s, r) {
  let a = _(e, "x", "avgPool", "float32"), i = 1;
  F(Ps(n, i), () => `Error in avgPool: Either strides or dilations must be 1. Got strides ${n} and dilations '${i}'`);
  let o = a, u = false;
  a.rank === 3 && (u = true, o = U(a, [1, a.shape[0], a.shape[1], a.shape[2]])), F(o.rank === 4, () => `Error in avgPool: x must be rank 4 but got rank ${o.rank}.`), hn("avgPool", s, r);
  let l = { x: o }, c = { filterSize: t, strides: n, pad: s, dimRoundingMode: r }, p = M.runKernel(Ca, l, c);
  return p = le(p, a.dtype), u ? U(p, [p.shape[1], p.shape[2], p.shape[3]]) : p;
}
var Qg = L({ avgPool_: CE });
function NE(e, t, n, s, r, a = "NDHWC") {
  let i = _(e, "x", "avgPool3d", "float32"), o = i, u = false;
  i.rank === 4 && (u = true, o = U(i, [1, i.shape[0], i.shape[1], i.shape[2], i.shape[3]])), F(o.rank === 5, () => `Error in avgPool3d: x must be rank 5 but got rank ${o.rank}.`), F(a === "NDHWC", () => `Error in avgPool3d: Only NDHWC is currently supported, but got dataFormat of ${a}`), hn("avgPool3d", s, r);
  let l = { x: o }, c = { filterSize: t, strides: n, pad: s, dimRoundingMode: r, dataFormat: a }, p = M.runKernel(Jd, l, c);
  return p = le(p, o.dtype), u ? U(p, [p.shape[1], p.shape[2], p.shape[3], p.shape[4]]) : p;
}
var iS = L({ avgPool3d_: NE });
function TE(e, t = 0) {
  F(e.length >= 1, () => "Pass at least one tensor to concat");
  let n = Ku(e, "tensors", "concat", "string_or_numeric");
  if (n[0].dtype === "complex64" && n.forEach((a) => {
    if (a.dtype !== "complex64")
      throw new Error(`Cannot concatenate complex64 tensors with a tensor
          with dtype ${a.dtype}. `);
  }), n.length === 1)
    return ur(n[0]);
  let s = n, r = { axis: t };
  return M.runKernel(fo, s, r);
}
var Ot = L({ concat_: TE });
function $E(e) {
  let n = { x: _(e, "x", "sigmoid", "float32") };
  return M.runKernel(ui, n);
}
var Hs = L({ sigmoid_: $E });
function _E(e, t, n) {
  let s = _(e, "x", "slice", "string_or_numeric");
  if (s.rank === 0)
    throw new Error("Slicing scalar is not possible");
  let r = { x: s }, a = { begin: t, size: n };
  return M.runKernel(Bo, r, a);
}
var qe = L({ slice_: _E });
function AE(e) {
  let n = { x: _(e, "x", "tanh", "float32") };
  return M.runKernel(fi, n);
}
var Yu = L({ tanh_: AE });
function EE(e, t, n, s, r, a) {
  let i = _(e, "forgetBias", "basicLSTMCell"), o = _(t, "lstmKernel", "basicLSTMCell"), u = _(n, "lstmBias", "basicLSTMCell"), l = _(s, "data", "basicLSTMCell"), c = _(r, "c", "basicLSTMCell"), p = _(a, "h", "basicLSTMCell"), d = Ot([l, p], 1), h = Ve(d, o), f = ie(h, u), m = f.shape[0], g = f.shape[1] / 4, b = [m, g], y = qe(f, [0, 0], b), v = qe(f, [0, g], b), x = qe(f, [0, g * 2], b), k = qe(f, [0, g * 3], b), I = ie(V(Hs(y), Yu(v)), V(c, Hs(ie(i, x)))), $ = V(Yu(I), Hs(k));
  return [I, $];
}
var Ipe = L({ basicLSTMCell_: EE });
function RE(e, t, n) {
  let s = _(e, "x", "batchToSpaceND"), r = t.reduce((o, u) => o * u);
  F(s.rank >= 1 + t.length, () => `input rank is ${s.rank} but should be > than blockShape.length ${t.length}`), F(n.length === t.length, () => `crops.length is ${n.length} but should be equal to blockShape.length  ${t.length}`), F(s.shape[0] % r === 0, () => `input tensor batch is ${s.shape[0]} but is not divisible by the product of the elements of blockShape ${t.join(" * ")} === ${r}`);
  let a = { x: s }, i = { blockShape: t, crops: n };
  return M.runKernel(ho, a, i);
}
var Zg = L({ batchToSpaceND_: RE });
function DE(e) {
  let t;
  return e.rank === 0 || e.rank === 1 ? t = U(e, [1, 1, 1, e.size]) : e.rank === 2 ? t = U(e, [1, 1, e.shape[0], e.shape[1]]) : e.rank === 3 ? t = U(e, [1, e.shape[0], e.shape[1], e.shape[2]]) : t = e, t;
}
function FE(e, t, n, s, r, a) {
  a == null && (a = 1e-3);
  let i = _(e, "x", "batchNorm"), o = _(t, "mean", "batchNorm"), u = _(n, "variance", "batchNorm"), l;
  r != null && (l = _(r, "scale", "batchNorm"));
  let c;
  s != null && (c = _(s, "offset", "batchNorm")), F(o.rank === u.rank, () => "Batch normalization gradient requires mean and variance to have equal ranks."), F(c == null || o.rank === c.rank, () => "Batch normalization gradient requires mean and offset to have equal ranks."), F(l == null || o.rank === l.rank, () => "Batch normalization gradient requires mean and scale to have equal ranks.");
  let d = { x: DE(i), scale: l, offset: c, mean: o, variance: u }, h = { varianceEpsilon: a }, f = M.runKernel(Ba, d, h);
  return U(f, i.shape);
}
var Qu = L({ batchNorm_: FE });
function OE(e, t, n, s, r, a) {
  let i = _(e, "x", "batchNorm"), o = _(t, "mean", "batchNorm"), u = _(n, "variance", "batchNorm"), l;
  r != null && (l = _(r, "scale", "batchNorm"));
  let c;
  return s != null && (c = _(s, "offset", "batchNorm")), F(i.rank === 2, () => `Error in batchNorm2D: x must be rank 2 but got rank ${i.rank}.`), F(o.rank === 2 || o.rank === 1, () => `Error in batchNorm2D: mean must be rank 2 or rank 1 but got rank ${o.rank}.`), F(u.rank === 2 || u.rank === 1, () => `Error in batchNorm2D: variance must be rank 2 or rank 1 but got rank ${u.rank}.`), l != null && F(l.rank === 2 || l.rank === 1, () => `Error in batchNorm2D: scale must be rank 2 or rank 1 but got rank ${l.rank}.`), c != null && F(c.rank === 2 || c.rank === 1, () => `Error in batchNorm2D: offset must be rank 2 or rank 1 but got rank ${c.rank}.`), Qu(i, o, u, c, l, a);
}
var PE = L({ batchNorm2d_: OE });
function zE(e, t, n, s, r, a) {
  let i = _(e, "x", "batchNorm"), o = _(t, "mean", "batchNorm"), u = _(n, "variance", "batchNorm"), l;
  r != null && (l = _(r, "scale", "batchNorm"));
  let c;
  return s != null && (c = _(s, "offset", "batchNorm")), F(i.rank === 3, () => `Error in batchNorm3D: x must be rank 3 but got rank ${i.rank}.`), F(o.rank === 3 || o.rank === 1, () => `Error in batchNorm3D: mean must be rank 3 or rank 1 but got rank ${o.rank}.`), F(u.rank === 3 || u.rank === 1, () => `Error in batchNorm3D: variance must be rank 3 or rank 1 but got rank ${u.rank}.`), l != null && F(l.rank === 3 || l.rank === 1, () => `Error in batchNorm3D: scale must be rank 3 or rank 1 but got rank ${l.rank}.`), c != null && F(c.rank === 3 || c.rank === 1, () => `Error in batchNorm3D: offset must be rank 3 or rank 1 but got rank ${c.rank}.`), Qu(i, o, u, c, l, a);
}
var ME = L({ batchNorm3d_: zE });
function LE(e, t, n, s, r, a) {
  let i = _(e, "x", "batchNorm"), o = _(t, "mean", "batchNorm"), u = _(n, "variance", "batchNorm"), l;
  r != null && (l = _(r, "scale", "batchNorm"));
  let c;
  return s != null && (c = _(s, "offset", "batchNorm")), F(i.rank === 4, () => `Error in batchNorm4D: x must be rank 4 but got rank ${i.rank}.`), F(o.rank === 4 || o.rank === 1, () => `Error in batchNorm4D: mean must be rank 4 or rank 1 but got rank ${o.rank}.`), F(u.rank === 4 || u.rank === 1, () => `Error in batchNorm4D: variance must be rank 4 or rank 1 but got rank ${u.rank}.`), l != null && F(l.rank === 4 || l.rank === 1, () => `Error in batchNorm4D: scale must be rank 4 or rank 1 but got rank ${l.rank}.`), c != null && F(c.rank === 4 || c.rank === 1, () => `Error in batchNorm4D: offset must be rank 4 or rank 1 but got rank ${c.rank}.`), Qu(i, o, u, c, l, a);
}
var BE = L({ batchNorm4d_: LE });
function VE(e, t, n) {
  let s = _(e, "x", "bincount"), r = _(t, "weights", "bincount");
  F(s.dtype === "int32", () => `Error in bincount: input dtype must be int32, but got ${s.dtype}`), F(n >= 0, () => `size must be non-negative, but got ${n}.`), F(r.size === s.size || r.size === 0, () => `Error in bincount: weights must have the same size as input or0-length, but got input shape: ${s.shape}, weights shape: ${r.shape}.`);
  let a = { x: s, weights: r }, i = { size: n };
  return M.runKernel(fg, a, i);
}
var oS = L({ bincount_: VE });
function WE(e, t) {
  let n = _(e, "s0", "broadcastArgs", "int32"), s = _(t, "s1", "broadcastArgs", "int32");
  if (n.rank !== 1)
    throw new Error(`broadcastArgs(): first input must be a vector (rank=1). Has rank ${n.rank}`);
  if (s.rank !== 1)
    throw new Error(`broadcastArgs(): second input must be a vector (rank=1). Has rank ${s.rank}`);
  let r = { s0: n, s1: s };
  return M.runKernel(mg, r);
}
var UE = L({ broadcastArgs_: WE });
function GE(e, t) {
  let n = _(e, "broadcastTo", "x"), s = n.shape;
  if (t.some((l) => !(l > 0) || l % 1 !== 0))
    throw new Error(`broadcastTo(): Invalid broadcast shape [${t}].`);
  if (t.length < n.rank)
    throw new Error(`broadcastTo(): shape.length=${t.length} < input.rank=${n.rank}.`);
  if (t.length > n.rank) {
    let l = n.shape.slice();
    for (; l.length < t.length; )
      l.unshift(1);
    n = U(n, l);
  }
  let r = n.shape, a = Array.from(t);
  for (let l = t.length - 1; l >= 0; l--)
    if (r[l] === t[l])
      a[l] = 1;
    else if (n.shape[l] !== 1)
      throw new Error(`broadcastTo(): [${s}] cannot be broadcast to [${t}].`);
  if (a.map((l, c) => l > 1 ? c : -1).filter((l) => l >= 0).length === 0)
    return ur(n);
  let o = { x: n }, u = { reps: a };
  return M.runKernel(Cr, o, u);
}
var ad = L({ broadcastTo_: GE });
function HE(e) {
  let n = { x: _(e, "x", "ceil", "float32") };
  return M.runKernel($a, n);
}
var qE = L({ ceil_: HE });
function jE(e, t, n) {
  let s = _(e, "x", "clipByValue");
  F(t <= n, () => `Error in clip: min (${t}) must be less than or equal to max (${n}).`);
  let r = { x: s }, a = { clipValueMin: t, clipValueMax: n };
  return M.runKernel(Ir, r, a);
}
var Vn = L({ clipByValue_: jE });
function KE(e) {
  return Ot(e, 0);
}
var XE = L({ concat1d_: KE });
function YE(e, t) {
  return Ot(e, t);
}
var QE = L({ concat2d_: YE });
function ZE(e, t) {
  return Ot(e, t);
}
var JE = L({ concat3d_: ZE });
function eR(e, t) {
  return Ot(e, t);
}
var tR = L({ concat4d_: eR });
function nR(e, t, n, s, r = "NHWC", a = [1, 1], i) {
  let o = _(e, "x", "conv2d", "float32"), u = _(t, "filter", "conv2d", "float32"), l = o, c = false;
  o.rank === 3 && (c = true, l = U(o, [1, o.shape[0], o.shape[1], o.shape[2]])), F(l.rank === 4, () => `Error in conv2d: input must be rank 4, but got rank ${l.rank}.`), F(u.rank === 4, () => `Error in conv2d: filter must be rank 4, but got rank ${u.rank}.`), hn("conv2d", s, i);
  let p = r === "NHWC" ? l.shape[3] : l.shape[1];
  F(p === u.shape[2], () => `Error in conv2d: depth of input (${p}) must match input depth for filter ${u.shape[2]}.`), F(Ps(n, a), () => `Error in conv2D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`);
  let d = { x: l, filter: u }, h = { strides: n, pad: s, dataFormat: r, dilations: a, dimRoundingMode: i }, f = M.runKernel(_a, d, h);
  return c ? U(f, [f.shape[1], f.shape[2], f.shape[3]]) : f;
}
var da = L({ conv2d_: nR });
function sR(e, t, n, s, r = "NWC", a = 1, i) {
  let o = _(e, "x", "conv1d"), u = _(t, "filter", "conv1d"), l = o, c = false;
  o.rank === 2 && (c = true, l = U(o, [1, o.shape[0], o.shape[1]])), F(l.rank === 3, () => `Error in conv1d: input must be rank 3, but got rank ${l.rank}.`), F(u.rank === 3, () => `Error in conv1d: filter must be rank 3, but got rank ${u.rank}.`), hn("conv1d", s, i), F(l.shape[2] === u.shape[1], () => `Error in conv1d: depth of input (${l.shape[2]}) must match input depth for filter ${u.shape[1]}.`), F(Ps(n, a), () => `Error in conv1D: Either stride or dilation must be 1. Got stride ${n} and dilation '${a}'`), F(r === "NWC", () => `Error in conv1d: got dataFormat of ${r} but only NWC is currently supported.`);
  let p = U(u, [1, u.shape[0], u.shape[1], u.shape[2]]), d = U(l, [l.shape[0], 1, l.shape[1], l.shape[2]]), g = da(d, p, [1, n], s, "NHWC", [1, a], i);
  return c ? U(g, [g.shape[2], g.shape[3]]) : U(g, [g.shape[0], g.shape[2], g.shape[3]]);
}
var uS = L({ conv1d_: sR });
function rR(e, t, n, s, r, a = "NHWC", i) {
  F(e.length === t.rank, () => `Length of inShape (${e.length}) and rank of dy (${t.rank}) must match`);
  let o = e, u = t, l = false;
  t.rank === 3 && (l = true, u = U(t, [1, t.shape[0], t.shape[1], t.shape[2]]), o = [1, e[0], e[1], e[2]]), F(o.length === 4, () => `Error in conv2dDerInput: inShape must be length 4, but got length ${o.length}.`), F(u.rank === 4, () => `Error in conv2dDerInput: dy must be rank 4, but got rank ${u.rank}`), F(n.rank === 4, () => `Error in conv2dDerInput: filter must be rank 4, but got rank ${n.rank}`);
  let c = a === "NHWC" ? o[3] : o[1], p = a === "NHWC" ? u.shape[3] : u.shape[1];
  F(c === n.shape[2], () => `Error in conv2dDerInput: depth of input (${c}) must match input depth for filter ${n.shape[2]}.`), F(p === n.shape[3], () => `Error in conv2dDerInput: depth of output (${p}) must match output depth for filter ${n.shape[3]}.`), hn("conv2dDerInput", r, i);
  let d = { dy: u, filter: n }, h = { strides: s, pad: r, dataFormat: a, dimRoundingMode: i, inputShape: o }, f = M.runKernel(Aa, d, h);
  return l ? U(f, [f.shape[1], f.shape[2], f.shape[3]]) : f;
}
var Jg = L({ conv2DBackpropInput_: rR });
function aR(e, t, n, s, r, a) {
  let i = _(e, "x", "conv2dTranspose"), o = _(t, "filter", "conv2dTranspose");
  return Jg(n, i, o, s, r, "NHWC", a);
}
var lS = L({ conv2dTranspose_: aR });
function iR(e, t, n, s, r = "NDHWC", a = [1, 1, 1]) {
  let i = _(e, "x", "conv3d"), o = _(t, "filter", "conv3d"), u = i, l = false;
  i.rank === 4 && (l = true, u = U(i, [1, i.shape[0], i.shape[1], i.shape[2], i.shape[3]])), F(u.rank === 5, () => `Error in conv3d: input must be rank 5, but got rank ${u.rank}.`), F(o.rank === 5, () => `Error in conv3d: filter must be rank 5, but got rank ${o.rank}.`), F(u.shape[4] === o.shape[3], () => `Error in conv3d: depth of input (${u.shape[4]}) must match input depth for filter ${o.shape[3]}.`), F(Ps(n, a), () => `Error in conv3D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`), F(r === "NDHWC", () => `Error in conv3d: got dataFormat of ${r} but only NDHWC is currently supported.`);
  let c = { x: u, filter: o }, p = { strides: n, pad: s, dataFormat: r, dilations: a }, d = M.runKernel(np, c, p);
  return l ? U(d, [d.shape[1], d.shape[2], d.shape[3], d.shape[4]]) : d;
}
var cS = L({ conv3d_: iR });
function oR(e, t, n, s, r) {
  F(e.length === t.rank, () => `Length of inShape (${e.length}) and rank of dy (${t.rank}) must match`);
  let a = e, i = t, o = false;
  t.rank === 4 && (o = true, i = U(t, [1, t.shape[0], t.shape[1], t.shape[2], t.shape[3]]), a = [1, e[0], e[1], e[2], e[3]]);
  let u = a[4], l = i.shape[4];
  F(a.length === 5, () => `Error in conv3dDerInput: inShape must be length 5, but got length ${a.length}.`), F(i.rank === 5, () => `Error in conv3dDerInput: dy must be rank 5, but got rank ${i.rank}`), F(n.rank === 5, () => `Error in conv3dDerInput: filter must be rank 5, but got rank ${n.rank}`), F(u === n.shape[3], () => `Error in conv3dDerInput: depth of input (${u}) must match input depth for filter ${n.shape[3]}.`), F(l === n.shape[4], () => `Error in conv3dDerInput: depth of output (${l}) must match output depth for filter ${n.shape[4]}.`);
  let c = { dy: i, filter: n }, p = { pad: r, strides: s, inputShape: a }, d = M.runKernel(yg, c, p);
  return o ? U(d, [d.shape[1], d.shape[2], d.shape[3], d.shape[4]]) : d;
}
var dS = L({ conv3DBackpropInput_: oR });
function uR(e, t, n, s, r) {
  let a = _(e, "x", "conv3dTranspose"), i = _(t, "filter", "conv3dTranspose");
  return dS(n, a, i, s, r);
}
var lR = L({ conv3dTranspose_: uR });
function cR(e) {
  let n = { x: _(e, "x", "cos", "float32") };
  return M.runKernel(Ea, n);
}
var eb = L({ cos_: cR });
function dR(e) {
  let n = { x: _(e, "x", "cosh", "float32") };
  return M.runKernel(Ra, n);
}
var pS = L({ cosh_: dR });
function pR(e, t = 0, n = false, s = false) {
  let a = { x: _(e, "x", "cumprod") }, i = { axis: t, exclusive: n, reverse: s };
  return M.runKernel(mo, a, i);
}
var xm = L({ cumprod_: pR });
function hR(e, t = 0, n = false, s = false) {
  let a = { x: _(e, "x", "cumsum") }, i = { axis: t, exclusive: n, reverse: s };
  return M.runKernel(Da, a, i);
}
var hS = L({ cumsum_: hR });
function fR(e, t, n, s = false) {
  let r = _(e, "x", "denseBincount"), a = _(t, "weights", "denseBincount");
  F(r.dtype === "int32", () => `Error in denseBincount: input dtype must be int32, but got ${r.dtype}`), F(r.rank <= 2, () => `Error in denseBincount: input must be at most rank 2, but got rank ${r.rank}.`), F(n >= 0, () => `size must be non-negative, but got ${n}.`), F(a.size === r.size || a.size === 0, () => `Error in denseBincount: weights must have the same shape as x or 0-length, but got x shape: ${r.shape}, weights shape: ${a.shape}.`);
  let i = { x: r, weights: a }, o = { size: n, binaryOutput: s };
  return M.runKernel(vg, i, o);
}
var mR = L({ denseBincount_: fR });
function gR(e, t, n = "NHWC") {
  let s = _(e, "x", "depthToSpace", "float32"), r = n === "NHWC" ? s.shape[1] : s.shape[2], a = n === "NHWC" ? s.shape[2] : s.shape[3], i = n === "NHWC" ? s.shape[3] : s.shape[1];
  F(t > 1, () => `blockSize should be > 1 for depthToSpace, but was: ${t}`), F(r * t >= 0, () => `Negative dimension size caused by overflow when multiplying
    ${r} and ${t}  for depthToSpace with input shape
    ${s.shape}`), F(a * t >= 0, () => `Negative dimension size caused by overflow when multiplying
    ${a} and ${t} for depthToSpace with input shape
        ${s.shape}`), F(i % (t * t) === 0, () => `Dimension size must be evenly divisible by ${t * t} but is ${i} for depthToSpace with input shape ${s.shape}`);
  let o = { x: s }, u = { blockSize: t, dataFormat: n };
  return M.runKernel(bo, o, u);
}
var bR = L({ depthToSpace_: gR });
function yR(e, t, n, s, r = "NHWC", a = [1, 1], i) {
  let o = _(e, "x", "depthwiseConv2d", "float32"), u = _(t, "filter", "depthwiseConv2d", "float32"), l = o, c = false;
  o.rank === 3 && (c = true, l = U(o, [1, o.shape[0], o.shape[1], o.shape[2]])), F(l.rank === 4, () => `Error in depthwiseConv2d: input must be rank 4, but got rank ${l.rank}.`), F(u.rank === 4, () => `Error in depthwiseConv2d: filter must be rank 4, but got rank ${u.rank}.`), F(l.shape[3] === u.shape[2], () => `Error in depthwiseConv2d: number of input channels (${l.shape[3]}) must match the inChannels dimension in filter ${u.shape[2]}.`), hn("depthwiseConv2d", s, i);
  let p = { x: l, filter: u }, d = { strides: n, pad: s, dataFormat: r, dilations: a, dimRoundingMode: i }, h = M.runKernel(Fa, p, d);
  return c ? U(h, [h.shape[1], h.shape[2], h.shape[3]]) : h;
}
var xp = L({ depthwiseConv2d_: yR });
function vR(e) {
  let n = { x: _(e, "x", "diag") };
  return M.runKernel(kg, n);
}
var Cpe = L({ diag_: vR });
function xR(e, t, n, s, r = [1, 1], a = "NHWC") {
  let i = _(e, "x", "dilation2d"), o = _(t, "filter", "dilation2d");
  F(i.rank === 3 || i.rank === 4, () => `Error in dilation2d: input must be rank 3 or 4, but got rank ${i.rank}.`), F(o.rank === 3, () => `Error in dilation2d: filter must be rank 3, but got rank ${o.rank}.`), F(a === "NHWC", () => `Error in dilation2d: Only NHWC is currently supported, but got dataFormat of ${a}`);
  let u = i, l = false;
  i.rank === 3 && (u = U(i, [1, i.shape[0], i.shape[1], i.shape[2]]), l = true);
  let c = { x: u, filter: o }, p = { strides: n, pad: s, dilations: r }, d = M.runKernel(sp, c, p);
  return l ? U(d, [d.shape[1], d.shape[2], d.shape[3]]) : d;
}
var wR = L({ dilation2d_: xR });
function kR(e, t) {
  let n = _(e, "a", "equal", "string_or_numeric"), s = _(t, "b", "equal", "string_or_numeric");
  [n, s] = vt(n, s), rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(yo, r);
}
var Xn = L({ equal_: kR });
function SR(e, t, n) {
  let s = _(t, "a", "where"), r = _(n, "b", "where"), a = _(e, "condition", "where", "bool"), i = rt(rt(a.shape, s.shape), r.shape), o = ad(a, i), u = ad(s, i), l = ad(r, i), c = { condition: o, t: u, e: l };
  return M.runKernel(Lo, c);
}
var vn = L({ where_: SR });
function IR(e) {
  let n = { x: _(e, "x", "zerosLike") };
  return M.runKernel(Xo, n);
}
var je = L({ zerosLike_: IR });
function CR(e, t) {
  let n = _(e, "a", "div"), s = _(t, "b", "div");
  [n, s] = vt(n, s);
  let r = xe(n, s), a = je(r), i = Xn(s, a);
  return vn(i, a, r);
}
var NR = L({ divNoNan_: CR });
function TR(e, t) {
  let n = _(e, "t1", "dot"), s = _(t, "t2", "dot");
  F((n.rank === 1 || n.rank === 2) && (s.rank === 1 || s.rank === 2), () => `Error in dot: inputs must all be rank 1 or 2, but got ranks ${n.rank} and ${s.rank}.`);
  let r = n.rank === 1 ? n.size : n.shape[1], a = s.rank === 1 ? s.size : s.shape[0];
  if (F(r === a, () => `Error in dot: inner dimensions of inputs must match, but got ${r} and ${a}.`), n.rank === 1 && s.rank === 1) {
    let i = U(n, [1, -1]), o = U(s, [-1, 1]), u = Ve(i, o);
    return U(u, []);
  } else if (n.rank === 1 && s.rank === 2) {
    let i = U(n, [1, -1]), o = U(s, [s.shape[0], s.shape[1]]), u = Ve(i, o);
    return U(u, [u.size]);
  } else if (n.rank === 2 && s.rank === 1) {
    let i = U(s, [-1, 1]), o = Ve(n, i);
    return U(o, [o.size]);
  } else {
    let i = U(s, [s.shape[0], s.shape[1]]);
    return Ve(n, i);
  }
}
var Npe = L({ dot_: TR });
function $R(e, ...t) {
  let n = t.map((r, a) => _(r, `tensors${a}`, "einsum")), s = { equation: e };
  return M.runKernel(rp, n, s);
}
var _R = L({ einsum_: $R });
function AR(e) {
  let n = { x: _(e, "x", "elu", "float32") };
  return M.runKernel(Pa, n);
}
var wp = L({ elu_: AR });
function ER(e) {
  let t = _(e, "x", "erf");
  F(t.dtype === "int32" || t.dtype === "float32", () => "Input dtype must be `int32` or `float32`."), t.dtype === "int32" && (t = le(t, "float32"));
  let n = { x: t };
  return M.runKernel(bl, n);
}
var RR = L({ erf_: ER });
function tb(e, t) {
  for (let n = 0; n < e.length; ++n)
    if (e[e.length - n - 1] !== t - 1 - n)
      return false;
  return true;
}
function fS(e, t, n) {
  let s = e.length + t.length, r = [], a = 0, i = 0;
  for (let o = 0; o < s; o++)
    n.indexOf(o) === -1 ? r.push(e[a++]) : r.push(t[i++]);
  return r;
}
function mS(e, t) {
  let n = [], s = e.length;
  for (let a = 0; a < s; a++)
    t.indexOf(a) === -1 && n.push(e[a]);
  let r = t.map((a) => e[a]);
  return [n, r];
}
function pa(e, t) {
  let n = t.map((s) => 1);
  return fS(e, n, t);
}
function DR(e, t, n) {
  F(tb(t, n), () => `${e} supports only inner-most axes for now. Got axes ${t} and rank-${n} input.`);
}
function gS(e, t) {
  if (tb(e, t))
    return null;
  let n = [];
  for (let s = 0; s < t; ++s)
    e.indexOf(s) === -1 && n.push(s);
  return e.forEach((s) => n.push(s)), n;
}
function nb(e) {
  return e.map((t, n) => [n, t]).sort((t, n) => t[1] - n[1]).map((t) => t[0]);
}
function FR(e, t) {
  let n = [];
  for (let s = t - e; s < t; ++s)
    n.push(s);
  return n;
}
function OR(e, t = null, n = false) {
  let r = { x: _(e, "x", "max") }, a = { reductionIndices: t, keepDims: n };
  return M.runKernel(Ha, r, a);
}
var As = L({ max_: OR });
function PR(e, t = null, n = false) {
  let r = { x: _(e, "x", "min") }, a = { axis: t, keepDims: n };
  return M.runKernel(Xa, r, a);
}
var wm = L({ min_: PR });
function zR(e, t) {
  let n = _(e, "base", "pow"), s = _(t, "exp", "pow");
  [n, s] = vt(n, s);
  let r = { a: n, b: s };
  return M.runKernel(ei, r);
}
var ha = L({ pow_: zR });
function we(e, t) {
  if ((Qt(e) && t !== "string" || Array.isArray(e)) && t !== "complex64")
    throw new Error("Error creating a new Scalar: value must be a primitive (number|boolean|string)");
  if (t === "string" && Qt(e) && !(e instanceof Uint8Array))
    throw new Error("When making a scalar from encoded string, the value must be `Uint8Array`.");
  return Nr(e, [], [], t);
}
function MR(e) {
  let n = { x: _(e, "x", "sqrt", "float32") };
  return M.runKernel(li, n);
}
var dn = L({ sqrt_: MR });
function LR(e) {
  let t = _(e, "x", "square"), n = {};
  return M.runKernel("Square", { x: t }, n);
}
var ct = L({ square_: LR });
function BR(e, t = null, n = false) {
  let s = _(e, "x", "sum");
  s.dtype === "bool" && (s = le(s, "int32"));
  let r = { x: s }, a = { axis: t, keepDims: n };
  return M.runKernel(ci, r, a);
}
var ve = L({ sum_: BR });
function VR(e, t = "euclidean", n = null, s = false) {
  e = _(e, "x", "norm");
  let r = bS(e, t, n), a = r.shape;
  if (s) {
    let i = ts(n, e.shape);
    a = pa(r.shape, i);
  }
  return U(r, a);
}
function bS(e, t, n = null) {
  if (e.rank === 0)
    return Lt(e);
  if (e.rank !== 1 && n === null)
    return bS(U(e, [-1]), t, n);
  if (e.rank === 1 || typeof n == "number" || Array.isArray(n) && n.length === 1) {
    if (t === 1)
      return ve(Lt(e), n);
    if (t === 1 / 0)
      return As(Lt(e), n);
    if (t === -1 / 0)
      return wm(Lt(e), n);
    if (t === "euclidean" || t === 2)
      return dn(ve(ha(Lt(e), we(2, "int32")), n));
    throw new Error(`Error in norm: invalid ord value: ${t}`);
  }
  if (Array.isArray(n) && n.length === 2) {
    if (t === 1)
      return As(ve(Lt(e), n[0]), n[1] - 1);
    if (t === 1 / 0)
      return As(ve(Lt(e), n[1]), n[0]);
    if (t === -1 / 0)
      return wm(ve(Lt(e), n[1]), n[0]);
    if (t === "fro" || t === "euclidean")
      return dn(ve(ct(e), n));
    throw new Error(`Error in norm: invalid ord value: ${t}`);
  }
  throw new Error(`Error in norm: invalid axis: ${n}`);
}
var sb = L({ norm_: VR });
function WR(e, t = null, n = false) {
  return sb(e, "euclidean", t, n);
}
var UR = L({ euclideanNorm_: WR });
function GR(e) {
  let n = { x: _(e, "x", "exp") };
  return M.runKernel(za, n);
}
var Yn = L({ exp_: GR });
function HR(e, t = 0) {
  let n = _(e, "x", "expandDims", "string_or_numeric");
  F(t <= n.rank, () => "Axis must be <= rank of the tensor");
  let s = { input: n }, r = { dim: t };
  return M.runKernel(vo, s, r);
}
var Pn = L({ expandDims_: HR });
function qR(e) {
  let n = { x: _(e, "x", "expm1") };
  return M.runKernel(xo, n);
}
var jR = L({ expm1_: qR });
function KR(e, t) {
  let n = _(e, "x", "tile", "string_or_numeric");
  F(n.rank === t.length, () => `Error in transpose: rank of input ${n.rank} must match length of reps ${t}.`);
  let s = { x: n }, r = { reps: t };
  return M.runKernel(Cr, s, r);
}
var hs = L({ tile_: KR });
function XR(e, t, n, s = "float32") {
  t == null && (t = e);
  let r = Ae([e, t], s), a = e <= t ? e : t;
  for (let o = 0; o < a; ++o)
    r.set(1, o, o);
  let i = U(r.toTensor(), [e, t]);
  if (n == null)
    return i;
  if (n.length === 1)
    return hs(Pn(i, 0), [n[0], 1, 1]);
  if (n.length === 2)
    return hs(Pn(Pn(i, 0), 0), [n[0], n[1], 1, 1]);
  if (n.length === 3)
    return hs(Pn(Pn(Pn(i, 0), 0), 0), [n[0], n[1], n[2], 1, 1]);
  throw new Error(`eye() currently supports only 1D and 2D batchShapes, but received ${n.length}D.`);
}
var yS = L({ eye_: XR });
function Ll(e, t, n) {
  let s = { shape: e, value: t, dtype: n };
  return M.runKernel(yl, {}, s);
}
function YR(e) {
  let n = { x: _(e, "x", "floor", "float32") };
  return M.runKernel(Ma, n);
}
var kp = L({ floor_: YR });
function QR(e, t, n = 0, s = 0) {
  let r = _(e, "x", "gather"), a = _(t, "indices", "gather", "int32"), i = { x: r, indices: a }, o = { axis: n, batchDims: s };
  return M.runKernel(ko, i, o);
}
var Zu = L({ gather_: QR });
function ZR(e, t) {
  let n = _(e, "a", "greater", "string_or_numeric"), s = _(t, "b", "greater", "string_or_numeric");
  [n, s] = vt(n, s), rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(Io, r);
}
var Un = L({ greater_: ZR });
function JR(e, t) {
  let n = _(e, "a", "greaterEqual", "string_or_numeric"), s = _(t, "b", "greaterEqual", "string_or_numeric");
  [n, s] = vt(n, s), rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(Va, r);
}
var Zo = L({ greaterEqual_: JR });
function eD(e) {
  let n = { input: _(e, "input", "imag") };
  return M.runKernel(ap, n);
}
var rb = L({ imag_: eD });
function tD(e) {
  let n = { x: _(e, "x", "isFinite") };
  return M.runKernel(vl, n);
}
var Tpe = L({ isFinite_: tD });
function nD(e) {
  let n = { x: _(e, "x", "isInf") };
  return M.runKernel(xl, n);
}
var $pe = L({ isInf_: nD });
function sD(e) {
  let n = { x: _(e, "x", "isNaN") };
  return M.runKernel(wl, n);
}
var rD = L({ isNaN_: sD });
function aD(e, t = 0.2) {
  let s = { x: _(e, "x", "leakyRelu") }, r = { alpha: t };
  return M.runKernel(Ua, s, r);
}
var ab = L({ leakyRelu_: aD });
function iD(e, t) {
  let n = _(e, "a", "less", "string_or_numeric"), s = _(t, "b", "less", "string_or_numeric");
  [n, s] = vt(n, s), rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(Co, r);
}
var vS = L({ less_: iD });
function oD(e, t) {
  let n = _(e, "a", "lessEqual", "string_or_numeric"), s = _(t, "b", "lessEqual", "string_or_numeric");
  [n, s] = vt(n, s), rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(No, r);
}
var Jo = L({ lessEqual_: oD });
function uD(e, t, n) {
  if (n <= 0)
    throw new Error("The number of values should be positive.");
  let s = { start: e, stop: t, num: n };
  return M.runKernel(Ng, {}, s);
}
function lD(e, t = 5, n = 1, s = 1, r = 0.5) {
  let a = _(e, "x", "localResponseNormalization");
  F(a.rank === 4 || a.rank === 3, () => `Error in localResponseNormalization: x must be rank 3 or 4 but got
               rank ${a.rank}.`), F(eo(t), () => `Error in localResponseNormalization: depthRadius must be an integer but got depthRadius ${t}.`);
  let i = a, o = false;
  a.rank === 3 && (o = true, i = U(a, [1, a.shape[0], a.shape[1], a.shape[2]]));
  let u = { x: i }, l = { depthRadius: t, bias: n, alpha: s, beta: r }, c = M.runKernel(op, u, l);
  return o ? U(c, [c.shape[1], c.shape[2], c.shape[3]]) : c;
}
var cD = L({ localResponseNormalization_: lD });
function dD(e) {
  let n = { x: _(e, "x", "log", "float32") };
  return M.runKernel(Ga, n);
}
var Qn = L({ log_: dD });
function pD(e) {
  let n = { x: _(e, "x", "log1p") };
  return M.runKernel(kl, n);
}
var ib = L({ log1p_: pD });
function _pe(e) {
  return F(hr(e), () => "The f passed in grad(f) must be a function"), (t, n) => {
    let s = _(t, "x", "tf.grad", "string_or_numeric"), r = n != null ? _(n, "dy", "tf.grad") : null;
    return M.tidy(() => {
      let { value: a, grads: i } = M.gradients(() => e(s), [s], r);
      return r != null && pn(a.shape, r.shape, "The shape of dy passed in grad(f)(x, dy) must match the shape returned by f(x)"), Sp(i), i[0];
    });
  };
}
function Ape(e) {
  return F(hr(e), () => "The f passed in grads(f) must be a function"), (t, n) => {
    F(Array.isArray(t), () => "The args passed in grads(f)(args) must be an array of `Tensor`s or `TensorLike`s");
    let s = Ku(t, "args", "tf.grads", "string_or_numeric"), r = n != null ? _(n, "dy", "tf.grads") : null;
    return M.tidy(() => {
      let { value: a, grads: i } = M.gradients(() => e(...s), s, r);
      return r != null && pn(a.shape, r.shape, "The shape of dy passed in grads(f)([x1,...], dy) must match the shape returned by f([x1,...])"), Sp(i), i;
    });
  };
}
function Epe(e) {
  return F(hr(e), () => "The f passed in valueAndGrad(f) must be a function"), (t, n) => {
    F(t instanceof et, () => "The x passed in valueAndGrad(f)(x) must be a tensor"), F(n == null || n instanceof et, () => "The dy passed in valueAndGrad(f)(x, dy) must be a tensor");
    let { grads: s, value: r } = M.gradients(() => e(t), [t], n);
    return Sp(s), { grad: s[0], value: r };
  };
}
function Rpe(e) {
  return F(hr(e), () => "The f passed in valueAndGrads(f) must be a function"), (t, n) => {
    F(Array.isArray(t) && t.every((r) => r instanceof et), () => "The args passed in valueAndGrads(f)(args) must be array of tensors"), F(n == null || n instanceof et, () => "The dy passed in valueAndGrads(f)(args, dy) must be a tensor");
    let s = M.gradients(() => e(...t), t, n);
    return n != null && pn(s.value.shape, n.shape, "The shape of dy passed in valueAndGrads(f)([x1,...], dy) must match the shape returned by f([x1,...])"), Sp(s.grads), s;
  };
}
function hD(e, t) {
  F(hr(e), () => "The f passed in variableGrads(f) must be a function"), F(t == null || Array.isArray(t) && t.every((l) => l instanceof xd), () => "The varList passed in variableGrads(f, varList) must be an array of variables");
  let n = t != null;
  if (!n) {
    t = [];
    for (let l in M.registeredVariables)
      t.push(M.registeredVariables[l]);
  }
  let s = n ? t.filter((l) => !l.trainable) : null, r = t.length;
  t = t.filter((l) => l.trainable), F(t.length > 0, () => `variableGrads() expects at least one of the input variables to be trainable, but none of the ${r} variables is trainable.`);
  let a = true, { value: i, grads: o } = M.gradients(e, t, null, a);
  F(o.some((l) => l != null), () => "Cannot find a connection between any variable and the result of the loss function y=f(x). Please make sure the operations that use variables are inside the function f passed to minimize()."), F(i.rank === 0, () => `The f passed in variableGrads(f) must return a scalar, but it returned a rank-${i.rank} tensor`);
  let u = {};
  return t.forEach((l, c) => {
    o[c] != null && (u[l.name] = o[c]);
  }), s != null && s.forEach((l) => u[l.name] = null), { value: i, grads: u };
}
function qs(e) {
  return M.customGrad(e);
}
function Sp(e) {
  if (e.filter((n) => n == null).length > 0)
    throw new Error(`Cannot compute gradient of y=f(x) with respect to x. Make sure that
    the f you passed encloses all operations that lead from x to y.`);
}
function fD(e) {
  let n = { x: _(e, "x", "neg") };
  return M.runKernel($o, n);
}
var kt = L({ neg_: fD });
function mD(e) {
  let n = { x: _(e, "x", "softplus") };
  return M.runKernel(El, n);
}
var Bl = L({ softplus_: mD });
function gD(e) {
  let t = _(e, "x", "logSigmoid");
  return qs((s) => ({ value: kt(Bl(kt(s))), gradFunc: (i) => V(i, Hs(kt(s))) }))(t);
}
var Dpe = L({ logSigmoid_: gD });
function bD(e, t) {
  let n = _(e, "a", "sub"), s = _(t, "b", "sub");
  [n, s] = vt(n, s);
  let r = { a: n, b: s };
  return M.runKernel(hi, r);
}
var ge = L({ sub_: bD });
function yD(e, t = -1) {
  let n = _(e, "logits", "logSoftmax");
  if (t === -1 && (t = n.rank - 1), t !== n.rank - 1)
    throw Error(`Log Softmax along a non-last dimension is not yet supported. Logits was rank ${n.rank} and axis was ${t}`);
  return qs((r, a) => {
    let o = As(r, t, true), u = ge(r, o), l = ge(le(u, "float32"), Qn(ve(Yn(u), t, true)));
    return a([l]), { value: l, gradFunc: (p, d) => {
      let [h] = d, f = true, m = Yn(h);
      return ge(p, V(ve(p, t, f), m));
    } };
  })(n);
}
var xS = L({ logSoftmax_: yD });
function vD(e, t = null, n = false) {
  let s = _(e, "x", "logSumExp"), r = ts(t, s.shape), a = As(s, r, true), i = ge(s, a), o = Yn(i), u = ve(o, r), l = Qn(u), c = ie(U(a, l.shape), l);
  if (n) {
    let p = pa(c.shape, r);
    return U(c, p);
  }
  return c;
}
var xD = L({ logSumExp_: vD });
function wD(e, t) {
  let n = _(e, "a", "logicalAnd", "bool"), s = _(t, "b", "logicalAnd", "bool");
  rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(To, r);
}
var Ds = L({ logicalAnd_: wD });
function kD(e) {
  let n = { x: _(e, "x", "logicalNot", "bool") };
  return M.runKernel(Sl, n);
}
var ob = L({ logicalNot_: kD });
function SD(e, t) {
  let n = _(e, "a", "logicalOr", "bool"), s = _(t, "b", "logicalOr", "bool");
  rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(ip, r);
}
var wS = L({ logicalOr_: SD });
function ID(e, t) {
  let n = _(e, "a", "logicalXor", "bool"), s = _(t, "b", "logicalXor", "bool");
  return rt(n.shape, s.shape), Ds(wS(e, t), ob(Ds(e, t)));
}
var Fpe = L({ logicalXor_: ID });
var Gc = 2147483648;
function CD(e, t, n = "left") {
  let s = _(e, "sortedSequence", "searchSorted"), r = _(t, "values", "searchSorted"), a = s.shape[s.shape.length - 1], i = r.shape[r.shape.length - 1], o = U(s, [-1, a]), u = U(r, [-1, i]);
  if (o.rank < 2)
    throw new Error("Sorted input argument must be at least 2-dimensional");
  if (o.shape[0] !== u.shape[0])
    throw new Error("Leading dimension of 'sortedSequence' and 'values' must match.");
  if (dt(u.shape) >= Gc)
    throw new Error(`values tensor size must less than ${Gc}`);
  if (o.shape[1] >= Gc)
    throw new Error(`trailing dim_size must less than ${Gc} for int32 output type, was ${o.shape[1]}`);
  let l = { sortedSequence: o, values: u }, c = { side: n };
  return M.runKernel(Fg, l, c);
}
var kS = L({ searchSorted_: CD });
function ND(e, t) {
  return kS(e, t, "left");
}
function TD(e, t, n, s, r) {
  let a = _(e, "x", "maxPool"), i = 1, o = a, u = false;
  a.rank === 3 && (u = true, o = U(a, [1, a.shape[0], a.shape[1], a.shape[2]])), F(o.rank === 4, () => `Error in maxPool: input must be rank 4 but got rank ${o.rank}.`), F(Ps(n, i), () => `Error in maxPool: Either strides or dilations must be 1. Got strides ${n} and dilations '${i}'`), hn("maxPool", s, r);
  let l = { x: o }, c = { filterSize: t, strides: n, pad: s, dimRoundingMode: r }, p = M.runKernel(ja, l, c);
  return u ? U(p, [p.shape[1], p.shape[2], p.shape[3]]) : p;
}
var ub = L({ maxPool_: TD });
function $D(e, t = [1, 1, 1], n, s, r, a = "NDHWC") {
  let i = _(e, "x", "maxPool3d"), o = i, u = false;
  i.rank === 4 && (u = true, o = U(i, [1, i.shape[0], i.shape[1], i.shape[2], i.shape[3]])), F(o.rank === 5, () => `Error in maxPool3d: x must be rank 5 but got rank ${o.rank}.`), F(a === "NDHWC", () => `Error in maxPool3d: Only NDHWC is currently supported, but got dataFormat of ${a}`), hn("maxPool3d", s, r);
  let l = { x: o }, c = { filterSize: t, strides: n, pad: s, dimRoundingMode: r, dataFormat: a }, p = M.runKernel(up, l, c);
  return u ? U(p, [p.shape[1], p.shape[2], p.shape[3], p.shape[4]]) : p;
}
var SS = L({ maxPool3d_: $D });
function _D(e, t, n, s, r = false) {
  let i = { x: _(e, "x", "maxPoolWithArgmax") }, o = { filterSize: t, strides: n, pad: s, includeBatchInIndex: r }, u = M.runKernel(Ag, i, o);
  return { result: u[0], indexes: u[1] };
}
var AD = L({ maxPoolWithArgmax_: _D });
function ED(e, t) {
  let n = _(e, "a", "maximum"), s = _(t, "b", "maximum");
  [n, s] = vt(n, s), n.dtype === "bool" && (n = le(n, "int32"), s = le(s, "int32")), rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(qa, r);
}
var $r = L({ maximum_: ED });
function RD(e, t = null, n = false) {
  let r = { x: _(e, "x", "mean") }, a = { axis: t, keepDims: n };
  return M.runKernel(Ka, r, a);
}
var It = L({ mean_: RD });
function $t(e, t = "float32") {
  if (t === "complex64") {
    let s = $t(e, "float32"), r = $t(e, "float32");
    return ua(s, r);
  }
  let n = Zd(dt(e), t);
  return M.makeTensor(n, e, t);
}
function Mn(e, t = "float32") {
  if (t === "complex64") {
    let s = Mn(e, "float32"), r = $t(e, "float32");
    return ua(s, r);
  }
  let n = ug(dt(e), t);
  return M.makeTensor(n, e, t);
}
function Ope(e, t, { indexing: n = "xy" } = {}) {
  if (n !== "xy" && n !== "ij")
    throw new TypeError(`${n} is not a valid third argument to meshgrid`);
  if (e === void 0)
    return [];
  let s = _(e, "x", "meshgrid", e instanceof et ? e.dtype : "float32");
  if (t === void 0)
    return [s];
  let r = _(t, "y", "meshgrid", t instanceof et ? t.dtype : "float32"), a = dt(s.shape), i = dt(r.shape);
  return n === "xy" ? (s = U(s, [1, -1]), r = U(r, [-1, 1]), [Ve(Mn([i, 1], s.dtype), s), Ve(r, Mn([1, a], r.dtype))]) : (s = U(s, [-1, 1]), r = U(r, [1, -1]), [Ve(s, Mn([1, i], s.dtype)), Ve(Mn([a, 1], r.dtype), r)]);
}
function DD(e, t) {
  let n = _(e, "a", "minimum"), s = _(t, "b", "minimum");
  [n, s] = vt(n, s), n.dtype === "bool" && (n = le(n, "int32"), s = le(s, "int32")), rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(Ya, r);
}
var Ip = L({ minimum_: DD });
function FD(e, t, n) {
  F(n === "reflect" || n === "symmetric", () => `Invalid mode. Mode must be either reflect or symmetric. Got ${n}.`);
  let s = _(e, "x", "mirrorPad");
  if (s.rank === 0)
    throw new Error("mirrorPad(scalar) is not defined. Pass non-scalar to mirrorPad");
  F(t.length === s.rank, () => `Padding doesn't match input. Must be ${s.rank}. Got ${t.length}.`);
  let r = n === "reflect" ? 1 : 0;
  for (let o = 0; o < s.rank; o++)
    F(t[o].length === 2, () => "Invalid number of paddings. Must be length of 2 each."), F(t[o][0] >= 0 && t[o][0] <= s.shape[o] - r && t[o][1] >= 0 && t[o][1] <= s.shape[o] - r, () => `Padding in dimension ${o} cannot be greater than or equal to ${s.shape[o] - r} or less than 0 for input of shape ${s.shape}`);
  let a = { paddings: t, mode: n }, i = { x: s };
  return M.runKernel(Qa, i, a);
}
var OD = L({ mirrorPad_: FD });
function PD(e, t) {
  let n = _(e, "a", "mod"), s = _(t, "b", "mod");
  [n, s] = vt(n, s);
  let r = { a: n, b: s };
  return M.runKernel(Il, r);
}
var zD = L({ mod_: PD });
function MD(e, t = null, n = false) {
  e = _(e, "x", "moments");
  let s = ts(t, e.shape), r = It(e, s, n), a = r.shape;
  n || (a = pa(r.shape, s));
  let i = ct(ge(le(e, "float32"), U(r, a))), o = It(i, s, n);
  return { mean: r, variance: o };
}
var lb = L({ moments_: MD });
function LD(e, t, n, s) {
  let r = _(t, "data", "multiRNNCell"), a = Ku(n, "c", "multiRNNCell"), i = Ku(s, "h", "multiRNNCell"), o = r, u = [];
  for (let p = 0; p < e.length; p++) {
    let d = e[p](o, a[p], i[p]);
    u.push(d[0]), u.push(d[1]), o = d[1];
  }
  let l = [], c = [];
  for (let p = 0; p < u.length; p += 2)
    l.push(u[p]), c.push(u[p + 1]);
  return [l, c];
}
var Ppe = L({ multiRNNCell_: LD });
function BD(e, t, n, s = false) {
  let r = _(e, "logits", "multinomial"), a = r.size, i = r.rank;
  if (a < 2)
    throw new Error(`Error in multinomial: you need at least 2 outcomes, but got ${a}.`);
  if (i > 2)
    throw new Error(`Rank of probabilities must be 1 or 2, but is ${i}`);
  n = n || Math.random();
  let u = { logits: i === 1 ? U(r, [1, -1]) : r }, l = { numSamples: t, seed: n, normalized: s }, c = M.runKernel(Eg, u, l);
  return i === 1 ? U(c, [c.size]) : c;
}
var VD = L({ multinomial_: BD });
function WD(e, t) {
  let n = _(e, "a", "notEqual", "string_or_numeric"), s = _(t, "b", "notEqual", "string_or_numeric");
  [n, s] = vt(n, s), rt(n.shape, s.shape);
  let r = { a: n, b: s };
  return M.runKernel(_o, r);
}
var Ju = L({ notEqual_: WD });
function UD(e) {
  let n = { x: _(e, "x", "onesLike") };
  return M.runKernel(Ro, n);
}
var Zn = L({ onesLike_: UD });
function GD(e, t) {
  let n = _(e, "v1", "outerProduct"), s = _(t, "v2", "outerProduct");
  F(n.rank === 1 && s.rank === 1, () => `Error in outerProduct: inputs must be rank 1, but got ranks ${n.rank} and ${s.rank}.`);
  let r = U(n, [-1, 1]), a = U(s, [1, -1]);
  return Ve(r, a);
}
var zpe = L({ outerProduct_: GD });
function HD(e, t, n = 0) {
  let s = _(e, "x", "pad");
  if (s.rank === 0)
    throw new Error("pad(scalar) is not defined. Pass non-scalar to pad");
  let r = { paddings: t, constantValue: n }, a = { x: s };
  return M.runKernel(Ja, a, r);
}
var bi = L({ pad_: HD });
function qD(e, t, n = 0) {
  return F(t.length === 2, () => "Invalid number of paddings. Must be length of 2."), bi(e, [t], n);
}
var Mpe = L({ pad1d_: qD });
function jD(e, t, n = 0) {
  return F(t.length === 2 && t[0].length === 2 && t[1].length === 2, () => "Invalid number of paddings. Must be length of 2 each."), bi(e, t, n);
}
var Lpe = L({ pad2d_: jD });
function KD(e, t, n = 0) {
  return F(t.length === 3 && t[0].length === 2 && t[1].length === 2 && t[2].length === 2, () => "Invalid number of paddings. Must be length of 2 each."), bi(e, t, n);
}
var Bpe = L({ pad3d_: KD });
function XD(e, t, n = 0) {
  return F(t.length === 4 && t[0].length === 2 && t[1].length === 2 && t[2].length === 2 && t[3].length === 2, () => "Invalid number of paddings. Must be length of 2 each."), bi(e, t, n);
}
var Vpe = L({ pad4d_: XD });
function YD(e, t, n) {
  let s = _(e, "x", "spaceToBatchND");
  F(s.rank >= 1 + t.length, () => `input rank ${s.rank} should be > than [blockShape] ${t.length}`), F(n.length === t.length, () => `paddings.shape[0] ${n.length} must be equal to [blockShape] ${t.length}`), F(s.shape.reduce((i, o, u) => u > 0 && u <= t.length ? i && (o + n[u - 1][0] + n[u - 1][1]) % t[u - 1] === 0 : i, true), () => `input spatial dimensions ${s.shape.slice(1)} with paddings ${n.toString()} must be divisible by blockShapes ${t.toString()}`);
  let r = { x: s }, a = { blockShape: t, paddings: n };
  return M.runKernel(Wo, r, a);
}
var cb = L({ spaceToBatchND_: YD });
function QD(e, t, n, s, r, a, i) {
  r == null && (r = [1, 1]), a == null && (a = 1), s === 0 && (s = "valid");
  let o = _(e, "x", "maxPool"), u = o, l = false;
  o.rank === 3 && (l = true, u = U(o, [1, o.shape[0], o.shape[1], o.shape[2]])), F(Ps(a, r), () => `Error in pool: Either strides or dilations must be 1. Got strides ${a} and dilations '${r}'`);
  let c = sS(u.shape, t, a, r, s), p = [c.dilationHeight, c.dilationWidth], d;
  s === "same" ? d = JD([c.filterHeight, c.filterWidth], p) : d = [[0, 0], [0, 0]];
  let h = p[0] === 1 && p[1] === 1, [f, m] = ZD([c.inHeight, c.inWidth], p, d), g = h ? s : "valid", b = h ? u : cb(u, p, f), v = (n === "avg" ? () => Qg(b, t, a, g, i) : () => ub(b, t, a, g, i))(), x = h ? v : Zg(v, p, m);
  return l ? U(x, [x.shape[1], x.shape[2], x.shape[3]]) : x;
}
function ZD(e, t, n) {
  let s = n.map((c) => c[0]), r = n.map((c) => c[1]), a = e.concat(s, r), i = t.map((c, p) => (c - a[p] % c) % c), o = r.map((c, p) => c + i[p]), u = t.map((c, p) => [s[p], o[p]]), l = t.map((c, p) => [0, i[p]]);
  return [u, l];
}
function JD(e, t) {
  let s = e.map((i, o) => i + (i - 1) * (t[o] - 1)).map((i) => i - 1), r = s.map((i) => Math.floor(i / 2)), a = s.map((i, o) => i - r[o]);
  return s.map((i, o) => [r[o], a[o]]);
}
var Wpe = L({ pool_: QD });
function e3(e, t) {
  let n = _(e, "x", "prelu"), s = _(t, "alpha", "prelu"), r = { x: n, alpha: s };
  return M.runKernel(ti, r);
}
var db = L({ prelu_: e3 });
function t3(e, t = null, n = false) {
  let s = _(e, "x", "prod");
  s.dtype === "bool" && (s = le(s, "int32"));
  let r = { x: s }, a = { axis: t, keepDims: n };
  return M.runKernel(ni, r, a);
}
var IS = L({ prod_: t3 });
function n3(e, t, n) {
  let s = dt(e), r = null;
  if (n == null || n === "float32")
    r = new Float32Array(s);
  else if (n === "int32")
    r = new Int32Array(s);
  else if (n === "bool")
    r = new Uint8Array(s);
  else
    throw new Error(`Unknown data type ${n}`);
  for (let a = 0; a < s; a++)
    r[a] = t();
  return M.makeTensor(r, e, n);
}
var Upe = L({ rand_: n3 });
var pb = wa(Xd());
var hb = class {
  constructor(e, t, n, s, r) {
    this.mean = e, this.stdDev = t, this.dtype = n, this.nextVal = NaN, this.truncated = s, this.truncated && (this.upper = this.mean + this.stdDev * 2, this.lower = this.mean - this.stdDev * 2);
    let a = r || Math.random();
    this.random = pb.alea(a.toString());
  }
  nextValue() {
    if (!isNaN(this.nextVal)) {
      let s = this.nextVal;
      return this.nextVal = NaN, s;
    }
    let e, t, n = false;
    for (; !n; ) {
      let s, r, a;
      do
        s = 2 * this.random() - 1, r = 2 * this.random() - 1, a = s * s + r * r;
      while (a >= 1 || a === 0);
      let i = Math.sqrt(-2 * Math.log(a) / a);
      e = this.mean + this.stdDev * s * i, t = this.mean + this.stdDev * r * i, (!this.truncated || this.isValidTruncated(e)) && (n = true);
    }
    return (!this.truncated || this.isValidTruncated(t)) && (this.nextVal = this.convertValue(t)), this.convertValue(e);
  }
  convertValue(e) {
    return this.dtype == null || this.dtype === "float32" ? e : Math.round(e);
  }
  isValidTruncated(e) {
    return e <= this.upper && e >= this.lower;
  }
};
var s3 = class {
  constructor(e, t, n, s) {
    this.alpha = e, this.beta = 1 / t, this.dtype = n;
    let r = s || Math.random();
    this.randu = pb.alea(r.toString()), this.randn = new hb(0, 1, n, false, this.randu()), e < 1 ? this.d = e + 2 / 3 : this.d = e - 1 / 3, this.c = 1 / Math.sqrt(9 * this.d);
  }
  nextValue() {
    let e, t, n, s, r, a;
    for (; ; ) {
      do
        s = this.randn.nextValue(), a = 1 + this.c * s;
      while (a <= 0);
      if (a *= a * a, e = s * s, t = 1 - 0.331 * e * e, n = 0.5 * e + this.d * (1 - a + Math.log(a)), r = this.randu(), r < t || Math.log(r) < n)
        break;
    }
    return a = 1 / this.beta * this.d * a, this.alpha < 1 && (a *= Math.pow(this.randu(), 1 / this.alpha)), this.convertValue(a);
  }
  convertValue(e) {
    return this.dtype === "float32" ? e : Math.round(e);
  }
};
var r3 = class {
  constructor(e = 0, t = 1, n, s) {
    if (this.canReturnFloat = () => this.dtype == null || this.dtype === "float32", this.min = e, this.range = t - e, this.dtype = n, s == null && (s = Math.random()), typeof s == "number" && (s = s.toString()), !this.canReturnFloat() && this.range <= 1)
      throw new Error(`The difference between ${e} - ${t} <= 1 and dtype is not float`);
    this.random = pb.alea(s);
  }
  convertValue(e) {
    return this.canReturnFloat() ? e : Math.round(e);
  }
  nextValue() {
    return this.convertValue(this.min + this.range * this.random());
  }
};
function a3(e, t, n = 1, s = "float32", r) {
  if (n == null && (n = 1), s == null && (s = "float32"), s !== "float32" && s !== "int32")
    throw new Error(`Unsupported data type ${s}`);
  let a = new s3(t, n, s, r), i = Ae(e, s);
  for (let o = 0; o < i.values.length; o++)
    i.values[o] = a.nextValue();
  return i.toTensor();
}
var Gpe = L({ randomGamma_: a3 });
function i3(e, t = 0, n = 1, s, r) {
  if (s != null && s === "bool")
    throw new Error(`Unsupported data type ${s}`);
  let a = new hb(t, n, s, false, r), i = Ae(e, s);
  for (let o = 0; o < i.values.length; o++)
    i.values[o] = a.nextValue();
  return i.toTensor();
}
var o3 = L({ randomNormal_: i3 });
function u3(e, t = 0, n = 1, s = "float32", r) {
  let a = Ae(e, s), i = new r3(t, n, null, r);
  for (let o = 0; o < a.values.length; o++)
    a.values[o] = i.nextValue();
  return a.toTensor();
}
var Vl = L({ randomUniform_: u3 });
function el(e, t, n = 1, s = "float32") {
  if (n === 0)
    throw new Error("Cannot have a step of zero");
  let r = { start: e, stop: t, step: n, dtype: s };
  return M.runKernel(Nl, {}, r);
}
function l3(e) {
  let n = { input: _(e, "input", "real") };
  return M.runKernel(lp, n);
}
var Cd = L({ real_: l3 });
function c3(e) {
  let n = { x: _(e, "x", "reciprocal") };
  return M.runKernel(Tl, n);
}
var d3 = L({ reciprocal_: c3 });
function p3(e) {
  let n = { x: _(e, "x", "relu") };
  return M.runKernel(si, n);
}
var Xs = L({ relu_: p3 });
function h3(e) {
  let n = { x: _(e, "x", "relu6") };
  return M.runKernel(ai, n);
}
var CS = L({ relu6_: h3 });
function f3(e, t) {
  let s = { x: _(e, "x", "reverse") }, r = { dims: t };
  return M.runKernel(Po, s, r);
}
var Jn = L({ reverse_: f3 });
function m3(e) {
  let t = _(e, "x", "reverse");
  return F(t.rank === 1, () => `Error in reverse1D: x must be rank 1 but got rank ${t.rank}.`), Jn(t, 0);
}
var Hpe = L({ reverse1d_: m3 });
function g3(e, t) {
  let n = _(e, "x", "reverse");
  return F(n.rank === 2, () => `Error in reverse2D: x must be rank 2 but got rank ${n.rank}.`), Jn(n, t);
}
var qpe = L({ reverse2d_: g3 });
function b3(e, t) {
  let n = _(e, "x", "reverse");
  return F(n.rank === 3, () => `Error in reverse3D: x must be rank 3 but got rank ${n.rank}.`), Jn(n, t);
}
var jpe = L({ reverse3d_: b3 });
function y3(e, t) {
  let n = _(e, "x", "reverse");
  return F(n.rank === 4, () => `Error in reverse4D: x must be rank 4 but got rank ${n.rank}.`), Jn(n, t);
}
var Kpe = L({ reverse4d_: y3 });
function v3(e) {
  let n = { x: _(e, "x", "round") };
  return M.runKernel(zo, n);
}
var NS = L({ round_: v3 });
function x3(e) {
  let n = { x: _(e, "x", "rsqrt", "float32") };
  return M.runKernel(ii, n);
}
var TS = L({ rsqrt_: x3 });
function w3(e) {
  let n = { x: _(e, "x", "selu") };
  return M.runKernel(_l, n);
}
var $S = L({ selu_: w3 });
function k3(e, t, n, s, r, a = [1, 1], i = "NHWC") {
  let o = _(e, "x", "separableConv2d"), u = _(t, "depthwiseFilter", "separableConv2d"), l = _(n, "pointwiseFilter", "separableConv2d"), c = o, p = false;
  if (o.rank === 3 && (p = true, c = U(o, [1, o.shape[0], o.shape[1], o.shape[2]])), i === "NCHW")
    throw new Error("separableConv2d currently does not support dataFormat NCHW; only NHWC is supported");
  F(c.rank === 4, () => `Error in separableConv2d: input must be rank 4, but got rank ${c.rank}.`), F(u.rank === 4, () => `Error in separableConv2d: depthwise filter must be rank 4, but got rank ${u.rank}.`), F(l.rank === 4, () => `Error in separableConv2d: pointwise filter must be rank 4, but got rank ${u.rank}.`), F(l.shape[0] === 1, () => `Error in separableConv2d: the first dimension of pointwise filter  must be 1, but got ${l.shape[0]}.`), F(l.shape[1] === 1, () => `Error in separableConv2d: the second dimension of pointwise filter must be 1, but got ${l.shape[1]}.`);
  let d = u.shape[2], h = u.shape[3];
  F(l.shape[2] === d * h, () => `Error in separableConv2d: the third dimension of pointwise filter must be ${d * h}, but got ${l.shape[2]}.`);
  let f = xp(c, u, s, r, i, a), g = da(f, l, 1, "valid", i);
  return p ? U(g, [g.shape[1], g.shape[2], g.shape[3]]) : g;
}
var S3 = L({ separableConv2d_: k3 });
async function I3(e, t) {
  let n = _(e, "x", "setdiff1d"), s = _(t, "y", "setdiff1d");
  F(n.dtype === s.dtype, () => `x and y should have the same dtype, but got x (${n.dtype}) and y (${s.dtype}).`), F(n.rank === 1, () => `x should be 1D tensor, but got x (${n.shape}).`), F(s.rank === 1, () => `y should be 1D tensor, but got y (${s.shape}).`);
  let r = await n.data(), a = await s.data(), i = new Set(a), o = 0;
  for (let c = 0; c < r.length; c++)
    i.has(r[c]) || o++;
  let u = new Wt([o], n.dtype), l = new Wt([o], "int32");
  for (let c = 0, p = 0; c < r.length; c++)
    i.has(r[c]) || (u.values[p] = r[c], l.values[p] = c, p++);
  return [u.toTensor(), l.toTensor()];
}
var C3 = I3;
function N3(e) {
  let n = { x: _(e, "x", "sign") };
  return M.runKernel(Al, n);
}
var T3 = L({ sign_: N3 });
function $3(e) {
  let n = { x: _(e, "x", "sin", "float32") };
  return M.runKernel(oi, n);
}
var _S = L({ sin_: $3 });
function _3(e) {
  let n = { x: _(e, "x", "sinh") };
  return M.runKernel(Vo, n);
}
var AS = L({ sinh_: _3 });
function A3(e, t, n) {
  let s = _(e, "x", "slice1d");
  return F(s.rank === 1, () => `slice1d expects a rank-1 tensor, but got a rank-${s.rank} tensor`), qe(s, [t], [n]);
}
var fb = L({ slice1d_: A3 });
function E3(e, t, n) {
  let s = _(e, "x", "slice2d");
  return F(s.rank === 2, () => `slice2d expects a rank-2 tensor, but got a rank-${s.rank} tensor`), qe(s, t, n);
}
var ES = L({ slice2d_: E3 });
function R3(e, t, n) {
  let s = _(e, "x", "slice3d");
  return F(s.rank === 3, () => `slice3d expects a rank-3 tensor, but got a rank-${s.rank} tensor`), qe(s, t, n);
}
var mb = L({ slice3d_: R3 });
function D3(e, t, n) {
  let s = _(e, "x", "slice4d");
  return F(s.rank === 4, () => `slice4d expects a rank-4 tensor, but got a rank-${s.rank} tensor`), qe(s, t, n);
}
var Nd = L({ slice4d_: D3 });
function F3(e, t = -1) {
  let n = _(e, "logits", "softmax", "float32");
  if (t === -1 && (t = n.rank - 1), t !== n.rank - 1)
    throw Error(`Softmax along a non-last dimension is not yet supported. Logits was rank ${n.rank} and dim was ${t}`);
  let s = { logits: n }, r = { dim: t };
  return M.runKernel(di, s, r);
}
var gb = L({ softmax_: F3 });
function O3(e) {
  F(e.dtype === "complex64", () => `The dtype for tf.spectral.fft() must be complex64 but got ${e.dtype}.`);
  let t = { input: e };
  return M.runKernel(Ig, t);
}
var bb = L({ fft_: O3 });
function P3(e) {
  F(e.dtype === "complex64", () => `The dtype for tf.spectral.ifft() must be complex64 but got ${e.dtype}.`);
  let t = { input: e };
  return M.runKernel(Cg, t);
}
var Td = L({ ifft_: P3 });
function z3(e) {
  let t = e.shape[e.shape.length - 1], n = e.size / t, s;
  if (t <= 2) {
    let r = U(e, [n, t]);
    s = Td(r);
  } else {
    let r = [n, 2 * (t - 1)], a = U(Cd(e), [n, t]), i = U(rb(e), [n, t]), o = Jn(qe(a, [0, 1], [n, t - 2]), 1), u = V(Jn(qe(i, [0, 1], [n, t - 2]), 1), we(-1)), l = Ot([a, o], 1), c = Ot([i, u], 1), p = U(ua(l, c), [r[0], r[1]]);
    s = Td(p);
  }
  if (s = Cd(s), e.rank === 3 && e.shape[0] !== 0) {
    let r = s, a = e.shape[0];
    s = U(s, [a, s.shape[0] / a, s.shape[1]]), r.dispose();
  }
  return s;
}
var RS = L({ irfft_: z3 });
function M3(e, t, n = 0) {
  let r = { x: _(e, "x", "split") }, a = { numOrSizeSplits: t, axis: n };
  return M.runKernel(Uo, r, a);
}
var Bn = L({ split_: M3 });
function L3(e, t) {
  F(e.dtype === "float32", () => `The dtype for rfft() must be real value but got ${e.dtype}`);
  let n = e.shape[e.shape.length - 1], s = e.size / n, r;
  if (t != null && t < n) {
    let f = e.shape.map((g) => 0), m = e.shape.map((g) => g);
    m[e.shape.length - 1] = t, r = qe(e, f, m), n = t;
  } else if (t != null && t > n) {
    let f = e.shape.map((m) => m);
    f[e.shape.length - 1] = t - n, r = Ot([e, $t(f)], e.shape.length - 1), n = t;
  } else
    r = e;
  let a = je(r), i = U(ua(r, a), [s, n]), o = bb(i), u = Math.floor(n / 2) + 1, l = Cd(o), c = rb(o), p = Bn(l, [u, n - u], l.shape.length - 1), d = Bn(c, [u, n - u], c.shape.length - 1), h = r.shape.slice();
  return h[r.shape.length - 1] = u, U(ua(p[0], d[0]), h);
}
var yb = L({ rfft_: L3 });
function B3(e, t) {
  let n = _(e, "a", "squaredDifference"), s = _(t, "b", "squaredDifference");
  [n, s] = vt(n, s), rt(n.shape, s.shape);
  let r = { a: n, b: s }, a = {};
  return M.runKernel(pi, r, a);
}
var DS = L({ squaredDifference_: B3 });
function V3(e, t) {
  let n = _(e, "x", "squeeze");
  return U(n, ek(n.shape, t).newShape);
}
var mr = L({ squeeze_: V3 });
function W3(e, t = 0) {
  let n = Ku(e, "tensors", "stack", "string_or_numeric");
  F(n.length >= 1, () => "Pass at least one tensor to tf.stack"), n.length > 0 && F(t <= n[0].rank, () => "Axis must be <= rank of the tensor");
  let s = n, r = { axis: t };
  return M.runKernel(Fo, s, r);
}
var es = L({ stack_: W3 });
function U3(e, t = 0) {
  let s = { x: _(e, "x", "step") }, r = { alpha: t };
  return M.runKernel(gi, s, r);
}
var Cp = L({ step_: U3 });
function G3(e, t, n, s, r = 0, a = 0, i = 0, o = 0, u = 0) {
  let c = { x: _(e, "x", "stridedSlice", "string_or_numeric") }, p = { begin: t, end: n, strides: s, beginMask: r, endMask: a, ellipsisMask: i, newAxisMask: o, shrinkAxisMask: u };
  return M.runKernel(Go, c, p);
}
var H3 = L({ stridedSlice_: G3 });
function q3(e) {
  let n = { x: _(e, "x", "tan", "float32") };
  return M.runKernel(Ho, n);
}
var j3 = L({ tan_: q3 });
function Zt(e, t) {
  ka(e);
  let n = Rs(e, t);
  if (n.length !== 1)
    throw new Error("tensor1d() requires values to be a flat/TypedArray");
  return Nr(e, null, n, t);
}
function Zi(e, t, n) {
  if (ka(e), t != null && t.length !== 2)
    throw new Error("tensor2d() requires shape to have two numbers");
  let s = Rs(e, n);
  if (s.length !== 2 && s.length !== 1)
    throw new Error("tensor2d() requires values to be number[][] or flat/TypedArray");
  if (s.length === 1 && t == null)
    throw new Error("tensor2d() requires shape to be provided when `values` are a flat/TypedArray");
  return Nr(e, t, s, n);
}
function Xpe(e, t, n) {
  if (ka(e), t != null && t.length !== 4)
    throw new Error("tensor4d() requires shape to have four numbers");
  let s = Rs(e, n);
  if (s.length !== 4 && s.length !== 1)
    throw new Error("tensor4d() requires values to be number[][][][] or flat/TypedArray");
  if (s.length === 1 && t == null)
    throw new Error("tensor4d() requires shape to be provided when `values` are a flat array");
  return Nr(e, t, s, n);
}
function Ype(e, t, n) {
  if (ka(e), t != null && t.length !== 5)
    throw new Error("tensor5d() requires shape to have five numbers");
  let s = Rs(e, n);
  if (s.length !== 5 && s.length !== 1)
    throw new Error("tensor5d() requires values to be number[][][][][] or flat/TypedArray");
  if (s.length === 1 && t == null)
    throw new Error("tensor5d() requires shape to be provided when `values` are a flat array");
  return Nr(e, t, s, n);
}
function Qpe(e, t, n) {
  if (ka(e), t != null && t.length !== 6)
    throw new Error("tensor6d() requires shape to have six numbers");
  let s = Rs(e, n);
  if (s.length !== 6 && s.length !== 1)
    throw new Error("tensor6d() requires values to be number[][][][][][] or flat/TypedArray");
  if (s.length === 1 && t == null)
    throw new Error("tensor6d() requires shape to be provided when `values` are a flat array");
  return t = t || s, Nr(e, t, s, n);
}
function K3(e, t = 1, n = true) {
  let s = _(e, "x", "topk");
  if (s.rank === 0)
    throw new Error("topk() expects the input to be of rank 1 or higher");
  let r = s.shape[s.shape.length - 1];
  if (t < 0)
    throw new Error(`'k' passed to topk() must be >= 0 but got ${t}`);
  if (t > r)
    throw new Error(`'k' passed to topk() must be <= the last dimension (${r}) but got ${t}`);
  let a = { x: s }, i = { k: t, sorted: n }, [o, u] = M.runKernel(qo, a, i);
  return { values: o, indices: u };
}
var X3 = L({ topk_: K3 });
function Y3(e, t = 0, n = 1, s, r) {
  if (s != null && s === "bool")
    throw new Error("Unsupported data type $ { dtype }");
  let a = new hb(t, n, s, true, r), i = Ae(e, s);
  for (let o = 0; o < i.values.length; o++)
    i.values[o] = a.nextValue();
  return i.toTensor();
}
var vb = L({ truncatedNormal_: Y3 });
function Q3(e, t = 0) {
  let n = _(e, "x", "unique", "string_or_numeric");
  F(n.rank > 0, () => "The input tensor must be at least 1D");
  let s = { x: n }, r = { axis: t }, [a, i] = M.runKernel(zg, s, r);
  return { values: a, indices: i };
}
var xx = L({ unique_: Q3 });
function Z3(e, t, n) {
  let s = _(e, "x", "unsortedSegmentSum"), r = _(t, "segmentIds", "unsortedSegmentSum", "int32");
  F(eo(n), () => "numSegments must be of dtype int");
  let a = { x: s, segmentIds: r }, i = { numSegments: n };
  return M.runKernel(mp, a, i);
}
var J3 = L({ unsortedSegmentSum_: Z3 });
function eF(e, t = 0) {
  let n = _(e, "x", "unstack", "string_or_numeric");
  F(t >= -n.shape.length && t < n.shape.length, () => `Axis = ${t} is not in [-${n.shape.length}, ${n.shape.length})`);
  let s = { value: n }, r = { axis: t };
  return M.runKernel(Ko, s, r);
}
var Fs = L({ unstack_: eF });
function tF(e, t) {
  return kS(e, t, "right");
}
function nF(e, t = true, n, s) {
  return M.makeVariable(e, t, n, s);
}
function FS(e, t) {
  let n = [];
  for (let a = 0; a < t.length; a++)
    t[a] && n.push(a);
  let s = Ae(e, "int32"), r = Ae([n.length, e.length], "int32");
  for (let a = 0; a < n.length; a++) {
    let i = s.indexToLoc(n[a]), o = a * e.length;
    r.values.set(i, o);
  }
  return r.toTensor();
}
async function sF(e) {
  let t = _(e, "condition", "whereAsync", "bool"), n = await t.data(), s = FS(t.shape, n);
  return e !== t && t.dispose(), s;
}
var OS = sF;
async function rF(e, t, n) {
  let s = _(e, "tensor", "boolMask"), r = _(t, "mask", "boolMask", "bool"), a = n == null ? 0 : n, i = r.rank, o = s.shape;
  F(i > 0, () => "mask cannot be scalar"), pn(o.slice(a, a + i), r.shape, "mask's shape must match the first K dimensions of tensor's shape,");
  let u = 1;
  for (let m = a; m < a + i; m++)
    u *= o[m];
  let l = o.slice(0, a).concat([u], o.slice(a + i)), c = U(s, l), p = U(r, [-1]), d = await OS(p), h = mr(d, [1]), f = Zu(c, h, a);
  return e !== s && s.dispose(), t !== r && r.dispose(), h.dispose(), c.dispose(), p.dispose(), d.dispose(), f;
}
var Zpe = rF;
function aF(e, t, n, s, r = true) {
  let a = _(e, "v", "movingAverage"), i = _(t, "x", "movingAverage"), o = _(n, "decay", "movingAverage");
  yk(a, i), F(kr(a.shape, i.shape), () => "Shape mismatch in v and x");
  let u = we(1), l = ge(u, o), c = V(ge(i, a), l);
  if (r) {
    F(s != null, () => "When using zeroDebias: true, step is required.");
    let p = _(s, "step", "movingAverage");
    c = xe(c, ge(u, ha(o, p)));
  }
  return ie(a, c);
}
var Jpe = L({ movingAverage_: aF });
function iF(e, t, n) {
  let s = _(e, "indices", "scatterND", "int32"), r = _(t, "updates", "scatterND");
  jg(r, s, n);
  let a = { indices: s, updates: r }, i = { shape: n };
  return M.runKernel(Mo, a, i);
}
var oF = L({ scatterND_: iF });
function uF(e, t, n, s) {
  if (e.dtype !== "int32")
    throw new Error(`tf.sparseToDense() expects the indices to be int32 type, but the dtype was ${e.dtype}.`);
  if (e.rank > 2)
    throw new Error(`sparseIndices should be a scalar, vector, or matrix, but got shape ${e.shape}.`);
  let r = e.rank > 0 ? e.shape[0] : 1, a = e.rank > 1 ? e.shape[1] : 1;
  if (n.length !== a)
    throw new Error(`outputShape has incorrect number of elements:, ${n.length}, should be: ${a}.`);
  let i = t.size;
  if (!(t.rank === 0 || t.rank === 1 && i === r))
    throw new Error(`sparseValues has incorrect shape ${t.shape}, should be [] or [${r}]`);
  if (t.dtype !== s.dtype)
    throw new Error("sparseValues.dtype must match defaultValues.dtype");
}
function lF(e, t, n, s = 0) {
  let r = _(e, "sparseIndices", "sparseToDense", "int32"), a = _(t, "sparseValues", "sparseToDense", "string_or_numeric"), i = _(s, "defaultValue", "sparseToDense", a.dtype);
  uF(r, a, n, i);
  let o = { sparseIndices: r, sparseValues: a, defaultValue: i }, u = { outputShape: n };
  return M.runKernel(hp, o, u);
}
var PS = L({ sparseToDense_: lF });
function cF(e, t) {
  let n = _(t, "indices", "gatherND", "int32"), r = { params: _(e, "x", "gatherND", "string_or_numeric"), indices: n };
  return M.runKernel(So, r);
}
var dF = L({ gatherND_: cF });
function pF(e, t) {
  if (t == null)
    return e.shape.slice();
  if (kr(e.shape, t))
    return t;
  if (e.shape.length === t.length) {
    let n = [];
    for (let s = 0; s < e.shape.length; s++)
      t[s] == null && e.shape[s] != null ? n.push(e.shape[s]) : n.push(t[s]);
    return n;
  }
  return t;
}
function hF(e, t, n, s) {
  let r = _(e, "x", "dropout");
  if (F(r.dtype === "float32", () => `x has to be a floating point tensor since it's going to be scaled, but got a ${r.dtype} tensor instead.`), F(t >= 0 && t < 1, () => `rate must be a float in the range [0, 1), but got ${t}.`), t === 0)
    return e instanceof et ? r.clone() : r;
  let a = pF(r, n), i = 1 - t, o = xe(kp(ie(Vl(a, 0, 1, "float32", s), i)), i);
  return V(r, o);
}
var fF = L({ dropout_: hF });
function mF(e) {
  return Math.floor(Math.pow(2, Math.ceil(Math.log(e) / Math.log(2))));
}
function zS(e, t, n) {
  let s = 1 - e % 2, r = new Float32Array(e);
  for (let a = 0; a < e; ++a) {
    let i = 2 * Math.PI * a / (e + s - 1);
    r[a] = t - n * Math.cos(i);
  }
  return Zt(r, "float32");
}
async function gF(e, t, n = 1) {
  let s = _(e, "predictions", "inTopK"), r = _(t, "targets", "inTopK");
  F(s.rank > 1, () => `inTopK() expects the predictions to be of rank 2 or higher, but got ${s.rank}`), F(s.rank - 1 === r.rank, () => `predictions rank should be 1 larger than targets rank, but got predictions rank ${s.rank} and targets rank ${r.rank}`), pn(s.shape.slice(0, s.shape.length - 1), r.shape, "predictions's shape should be align with the targets' shape, except the last dimension.");
  let a = s.shape[s.shape.length - 1];
  F(n > 0 && n <= a, () => `'k' passed to inTopK() must be > 0 && <= the predictions last dimension (${a}), but got ${n}`);
  let i = await s.data(), o = await r.data(), [u, l] = [i.length / a, a], c = tk("bool", u);
  for (let p = 0; p < u; p++) {
    let d = p * l, h = i.subarray(d, d + l), f = [];
    for (let m = 0; m < h.length; m++)
      f.push({ value: h[m], index: m });
    f.sort((m, g) => g.value - m.value), c[p] = 0;
    for (let m = 0; m < n; m++)
      if (f[m].index === o[p]) {
        c[p] = 1;
        break;
      }
  }
  return e !== s && s.dispose(), t !== r && r.dispose(), ms(c, r.shape, "bool");
}
var ehe = gF;
var fa = {};
Ee(fa, { conv2d: () => vF, depthwiseConv2d: () => SF, matMul: () => CF });
function bF(e, t, n, s, r, a = "NHWC", i) {
  let o = e;
  e.rank === 3 && (o = U(e, [1, e.shape[0], e.shape[1], e.shape[2]]));
  let u = t;
  u.rank === 3 && (u = U(t, [1, t.shape[0], t.shape[1], t.shape[2]])), F(o.rank === 4, () => `Error in conv2dDerFilter: input must be rank 4, but got shape ${o.shape}.`), F(u.rank === 4, () => `Error in conv2dDerFilter: dy must be rank 4, but got shape ${u.shape}.`), F(n.length === 4, () => `Error in conv2dDerFilter: filterShape must be length 4, but got ${n}.`);
  let l = a === "NHWC" ? o.shape[3] : o.shape[1], c = a === "NHWC" ? u.shape[3] : u.shape[1];
  F(l === n[2], () => `Error in conv2dDerFilter: depth of input ${l}) must match input depth in filter (${n[2]}.`), F(c === n[3], () => `Error in conv2dDerFilter: depth of dy (${c}) must match output depth for filter (${n[3]}).`), hn("conv2dDerFilter", r, i);
  let p = { x: o, dy: u }, d = { strides: s, pad: r, dataFormat: a, dimRoundingMode: i, filterShape: n };
  return M.runKernel(gg, p, d);
}
var xb = L({ conv2DBackpropFilter_: bF });
function Np(e, t, n) {
  if (n == null || n === "linear")
    return e;
  if (n === "relu")
    return V(e, Cp(t));
  throw new Error(`Cannot compute gradient for fused activation ${n}.`);
}
function Tp(e, t) {
  let n = t, s = At(e.shape, t.shape);
  return s.length > 0 && (n = ve(n, s)), U(n, e.shape);
}
function $p(e, t, n, s) {
  if (t === "linear")
    return e;
  if (t === "relu")
    return Xs(e);
  if (t === "elu")
    return wp(e);
  if (t === "relu6")
    return CS(e);
  if (t === "prelu")
    return db(e, n);
  if (t === "leakyrelu")
    return ab(e, s);
  if (t === "sigmoid")
    return Hs(e);
  throw new Error(`Unknown fused activation ${t}.`);
}
var _p = (e, t) => !(e > 0) || t === "linear";
function yF({ x: e, filter: t, strides: n, pad: s, dataFormat: r = "NHWC", dilations: a = [1, 1], dimRoundingMode: i, bias: o, activation: u = "linear", preluActivationWeights: l, leakyreluAlpha: c }) {
  if (u = u || "linear", _p(M.state.gradientDepth, u) === false) {
    F(r === "NHWC", () => `Error in fused conv2d: got dataFormat of ${r} but only NHWC is currently supported for the case of gradient depth is 0 and the activation is not linear.`);
    let I = da(e, t, n, s, r, a, i);
    return o != null && (I = ie(I, o)), $p(I, u, l, c);
  }
  let p = _(e, "x", "conv2d", "float32"), d = _(t, "filter", "conv2d", "float32"), h = p, f = false;
  p.rank === 3 && (f = true, h = U(p, [1, p.shape[0], p.shape[1], p.shape[2]])), F(h.rank === 4, () => `Error in fused conv2d: input must be rank 4, but got rank ${h.rank}.`), F(d.rank === 4, () => `Error in fused conv2d: filter must be rank 4, but got rank ${d.rank}.`), hn("fused conv2d", s, i);
  let m = r === "NHWC" ? h.shape[3] : h.shape[1];
  F(d.shape[2] === m, () => `Error in conv2d: depth of input (${m}) must match input depth for filter ${d.shape[2]}.`), F(Ps(n, a), () => `Error in conv2D: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`);
  let g = Ml(h.shape, d.shape, n, a, s, i), b;
  o != null && (b = _(o, "bias", "fused conv2d"), [b] = vt(b, p), r === "NHWC" ? rt(g.outShape, b.shape) : (F(b.shape.length <= 1, () => `Error in fused conv2d: only supports scalar or 1-D Tensor bias for NCHW format but got the bias of rank-${b.shape.length}.`), F(b.shape.length === 0 || b.shape[0] === g.outChannels || b.shape[0] === 1, () => `Error in fused conv2d: bias shape (${b.shape}) is not compatible with the number of output channels (${g.outChannels})`)));
  let y;
  if (l != null) {
    let I = l.shape;
    if (F(I.length <= 1 || I.length === 3, () => `Error in fused conv2d: only supports scalar, 1-D Tensor or 3-D Tensor PReLU activation weights but got a tensor of rank-${I.length}.`), I.length === 1)
      F(I[0] === 1 || I[0] === g.outChannels, () => `Error in fused conv2d: PReLU activation weights (${I}) is not compatible with the number of output channels (${g.outChannels}).`);
    else if (I.length === 3)
      try {
        rt(I, g.outShape);
      } catch ($) {
        let R = `Error in fused conv2d: PReLU activation weights (${I}) is not compatible with the output shape of the conv2d (${g.outShape}).`;
        throw Error(R);
      }
    y = _(l, "prelu weights", "fused conv2d");
  }
  let v = (I, $) => {
    F(r === "NHWC", () => `Error in gradient of fused conv2D: got dataFormat of ${r} but only NHWC is currently supported.`);
    let [R, E, P, A] = $, O = Np(I, P, u);
    F(fr(a), () => `Error in gradient of fused conv2D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${a}'`);
    let T = Jg(E.shape, O, R, n, s), z = xb(E, O, R.shape, n, s), W = [T, z];
    if (A != null) {
      let q = Tp(A, O);
      W.push(q);
    }
    return W;
  }, x = { x: h, filter: d, bias: b, preluActivationWeights: y }, k = { strides: n, pad: s, dataFormat: r, dilations: a, dimRoundingMode: i, activation: u, leakyreluAlpha: c };
  return o == null ? qs(($, R, E) => {
    let P = M.runKernel(ia, x, k);
    return E([R, $, P]), f && (P = U(P, [P.shape[1], P.shape[2], P.shape[3]])), { value: P, gradFunc: v };
  })(h, d) : qs(($, R, E, P) => {
    let A = M.runKernel(ia, x, k);
    return P([R, $, A, E]), f && (A = U(A, [A.shape[1], A.shape[2], A.shape[3]])), { value: A, gradFunc: v };
  })(h, d, b);
}
var vF = L({ fusedConv2d_: yF });
function xF(e, t, n, s, r, a = [1, 1], i) {
  let o = e;
  e.rank === 3 && (o = U(e, [1, e.shape[0], e.shape[1], e.shape[2]]));
  let u = t;
  u.rank === 3 && (u = U(t, [1, t.shape[0], t.shape[1], t.shape[2]]));
  let l = { x: o, dy: u }, c = { strides: s, pad: r, dimRoundingMode: i, dilations: a, filterShape: n };
  return M.runKernel(xg, l, c);
}
var MS = L({ depthwiseConv2dNativeBackpropFilter_: xF });
function wF(e, t, n, s, r, a = [1, 1], i) {
  let o = t, u = false;
  t.rank === 3 && (u = true, o = U(t, [1, t.shape[0], t.shape[1], t.shape[2]]));
  let l = { dy: o, filter: n }, c = { strides: s, pad: r, dimRoundingMode: i, dilations: a, inputShape: e }, p = M.runKernel(wg, l, c);
  return u ? U(p, [p.shape[1], p.shape[2], p.shape[3]]) : p;
}
var LS = L({ depthwiseConv2dNativeBackpropInput_: wF });
function kF({ x: e, filter: t, strides: n, pad: s, dataFormat: r = "NHWC", dilations: a = [1, 1], dimRoundingMode: i, bias: o, activation: u = "linear", preluActivationWeights: l, leakyreluAlpha: c }) {
  if (_p(M.state.gradientDepth, u) === false) {
    let k = xp(e, t, n, s, r, a, i);
    return o != null && (k = ie(k, o)), $p(k, u, l, c);
  }
  let p = _(e, "x", "depthwiseConv2d", "float32"), d = _(t, "filter", "depthwiseConv2d", "float32"), h = p, f = false;
  p.rank === 3 && (f = true, h = U(p, [1, p.shape[0], p.shape[1], p.shape[2]])), F(h.rank === 4, () => `Error in fused depthwiseConv2d: input must be rank 4, but got rank ${h.rank}.`), F(d.rank === 4, () => `Error in fused depthwiseConv2d: filter must be rank 4, but got rank ${d.rank}.`), F(h.shape[3] === d.shape[2], () => `Error in fused depthwiseConv2d: number of input channels (${h.shape[3]}) must match the inChannels dimension in filter ${d.shape[2]}.`), a == null && (a = [1, 1]), F(Ps(n, a), () => `Error in fused depthwiseConv2d: Either strides or dilations must be 1. Got strides ${n} and dilations '${a}'`), hn("fused depthwiseConv2d", s, i);
  let m = Ml(h.shape, d.shape, n, a, s, i, true), g;
  o != null && (g = _(o, "bias", "fused conv2d"), [g] = vt(g, p), rt(m.outShape, g.shape));
  let b;
  l != null && (b = _(l, "prelu weights", "fused depthwiseConv2d"));
  let y = (k, I) => {
    F(fr(a), () => `Error in gradient of fused depthwiseConv2d: dilation rates greater than 1 are not yet supported. Got dilations '${a}'`);
    let [$, R, E, P] = I, A = Np(k, E, u), O = LS(R.shape, A, $, n, s, a, i), T = MS(R, A, $.shape, n, s, a, i);
    if (P != null) {
      let z = Tp(g, A);
      return [O, T, z];
    }
    return [O, T];
  }, v = { x: h, filter: d, bias: g, preluActivationWeights: b }, x = { strides: n, pad: s, dataFormat: r, dilations: a, dimRoundingMode: i, activation: u, leakyreluAlpha: c };
  return o == null ? qs((I, $, R) => {
    let E = M.runKernel(oa, v, x);
    return R([$, I, E]), f && (E = U(E, [E.shape[1], E.shape[2], E.shape[3]])), { value: E, gradFunc: y };
  })(h, d) : qs((I, $, R, E) => {
    let P = M.runKernel(oa, v, x);
    return E([$, I, P, R]), f && (P = U(P, [P.shape[1], P.shape[2], P.shape[3]])), { value: P, gradFunc: y };
  })(h, d, g);
}
var SF = L({ fusedDepthwiseConv2d_: kF });
function IF({ a: e, b: t, transposeA: n = false, transposeB: s = false, bias: r, activation: a = "linear", preluActivationWeights: i, leakyreluAlpha: o }) {
  if (_p(M.state.gradientDepth, a) === false) {
    let A = Ve(e, t, n, s);
    return r != null && (A = ie(A, r)), $p(A, a, i, o);
  }
  let u = _(e, "a", "fused matMul"), l = _(t, "b", "fused matMul");
  [u, l] = vt(u, l);
  let c = n ? u.shape[u.rank - 2] : u.shape[u.rank - 1], p = s ? l.shape[l.rank - 1] : l.shape[l.rank - 2], d = n ? u.shape[u.rank - 1] : u.shape[u.rank - 2], h = s ? l.shape[l.rank - 2] : l.shape[l.rank - 1], f = u.shape.slice(0, -2), m = l.shape.slice(0, -2), g = dt(f), b = dt(m);
  F(c === p, () => `Error in fused matMul: inner shapes (${c}) and (${p}) of Tensors with shapes ${u.shape} and ${l.shape} and transposeA=${n} and transposeB=${s} must match.`);
  let v = rt(u.shape.slice(0, -2), l.shape.slice(0, -2)).concat([d, h]), x = n ? U(u, [g, c, d]) : U(u, [g, d, c]), k = s ? U(l, [b, h, p]) : U(l, [b, p, h]), I;
  r != null && (I = _(r, "bias", "fused matMul"), [I] = vt(I, u), rt(v, I.shape));
  let $;
  i != null && ($ = _(i, "prelu weights", "fused matMul"));
  let R = (A, O) => {
    let [T, z, W, q] = O, X = Np(U(A, W.shape), W, a), Y, Z;
    if (!n && !s ? (Y = Ve(X, z, false, true), Z = Ve(T, X, true, false)) : !n && s ? (Y = Ve(X, z, false, false), Z = Ve(X, T, true, false)) : n && !s ? (Y = Ve(z, X, false, true), Z = Ve(T, X, false, false)) : (Y = Ve(z, X, true, true), Z = Ve(X, T, true, true)), r != null) {
      let te = Tp(q, X);
      return [Y, Z, te];
    } else
      return [Y, Z];
  }, E = { a: x, b: k, bias: I, preluActivationWeights: $ }, P = { transposeA: n, transposeB: s, activation: a, leakyreluAlpha: o };
  return r == null ? qs((O, T, z) => {
    let W = M.runKernel(aa, E, P);
    return z([O, T, W]), { value: U(W, v), gradFunc: R };
  })(x, k) : qs((O, T, z, W) => {
    let q = M.runKernel(aa, E, P);
    return W([O, T, q, z]), { value: U(q, v), gradFunc: R };
  })(x, k, I);
}
var CF = L({ fusedMatMul_: IF });
function NF(e) {
  return zS(e, 0.54, 0.46);
}
var TF = L({ hammingWindow_: NF });
function $F(e) {
  return zS(e, 0.5, 0.5);
}
var BS = L({ hannWindow_: $F });
function _F(e, t, n, s = false, r = 0) {
  let a = 0, i = [];
  for (; a + t <= e.size; )
    i.push(qe(e, a, t)), a += n;
  if (s)
    for (; a < e.size; ) {
      let o = a + t - e.size, u = Ot([qe(e, a, t - o), Ll([o], r)]);
      i.push(u), a += n;
    }
  return i.length === 0 ? Zi([], [0, t]) : U(Ot(i), [i.length, t]);
}
var VS = L({ frame_: _F });
function AF(e, t, n, s, r = BS) {
  s == null && (s = mF(t));
  let a = VS(e, t, n), i = V(a, r(t));
  return yb(i, s);
}
var EF = L({ stft_: AF });
function RF(e, t, n, s, r = "bilinear", a = 0) {
  let i = _(e, "image", "cropAndResize"), o = _(t, "boxes", "cropAndResize", "float32"), u = _(n, "boxInd", "cropAndResize", "int32"), l = o.shape[0];
  F(i.rank === 4, () => `Error in cropAndResize: image must be rank 4,but got rank ${i.rank}.`), F(o.rank === 2 && o.shape[1] === 4, () => `Error in cropAndResize: boxes must be have size [${l},4] but had shape ${o.shape}.`), F(u.rank === 1 && u.shape[0] === l, () => `Error in cropAndResize: boxInd must be have size [${l}] but had shape ${o.shape}.`), F(s.length === 2, () => `Error in cropAndResize: cropSize must be of length 2, but got length ${s.length}.`), F(s[0] >= 1 && s[1] >= 1, () => `cropSize must be atleast [1,1], but was ${s}`), F(r === "bilinear" || r === "nearest", () => `method must be bilinear or nearest, but was ${r}`);
  let c = { image: i, boxes: o, boxInd: u }, p = { method: r, extrapolationValue: a, cropSize: s };
  return M.runKernel(go, c, p);
}
var DF = L({ cropAndResize_: RF });
function FF(e) {
  let t = _(e, "image", "flipLeftRight", "float32");
  F(t.rank === 4, () => `Error in flipLeftRight: image must be rank 4,but got rank ${t.rank}.`);
  let n = { image: t };
  return M.runKernel(wo, n, {});
}
var OF = L({ flipLeftRight_: FF });
function PF(e) {
  let t = _(e, "image", "grayscaleToRGB"), n = t.rank - 1, s = t.shape[n];
  F(t.rank >= 2, () => `Error in grayscaleToRGB: images must be at least rank 2, but got rank ${t.rank}.`), F(s === 1, () => `Error in grayscaleToRGB: last dimension of a grayscale image should be size 1, but got size ${s}.`);
  let r = new Array(t.rank);
  return r.fill(1, 0, n), r[n] = 3, hs(t, r);
}
var zF = L({ grayscaleToRGB_: PF });
function MF(e, t, n = 0, s = 0.5) {
  let r = _(e, "image", "rotateWithOffset", "float32");
  F(r.rank === 4, () => `Error in rotateWithOffset: image must be rank 4,but got rank ${r.rank}.`);
  let a = { image: r }, i = { radians: t, fillValue: n, center: s };
  return M.runKernel(Yo, a, i);
}
var LF = L({ rotateWithOffset_: MF });
function eu(e, t, n, s, r, a) {
  s == null && (s = 0.5), r == null && (r = Number.NEGATIVE_INFINITY), a == null && (a = 0);
  let i = e.shape[0];
  return n = Math.min(n, i), F(0 <= s && s <= 1, () => `iouThreshold must be in [0, 1], but was '${s}'`), F(e.rank === 2, () => `boxes must be a 2D tensor, but was of rank '${e.rank}'`), F(e.shape[1] === 4, () => `boxes must have 4 columns, but 2nd dimension was ${e.shape[1]}`), F(t.rank === 1, () => "scores must be a 1D tensor"), F(t.shape[0] === i, () => `scores has incompatible shape with boxes. Expected ${i}, but was ${t.shape[0]}`), F(0 <= a && a <= 1, () => `softNmsSigma must be in [0, 1], but was '${a}'`), { maxOutputSize: n, iouThreshold: s, scoreThreshold: r, softNmsSigma: a };
}
function BF(e, t, n, s = 0.5, r = Number.NEGATIVE_INFINITY) {
  let a = _(e, "boxes", "nonMaxSuppression", "float32"), i = _(t, "scores", "nonMaxSuppression", "float32"), o = eu(a, i, n, s, r);
  n = o.maxOutputSize, s = o.iouThreshold, r = o.scoreThreshold;
  let u = { maxOutputSize: n, iouThreshold: s, scoreThreshold: r };
  return M.runKernel(Ao, { boxes: a, scores: i }, u);
}
var VF = L({ nonMaxSuppression_: BF });
function WF(e, t, n) {
  let s = UF(e, t, n), r = s < 0 ? -(s + 1) : s;
  e.splice(r, 0, t);
}
function UF(e, t, n) {
  return HF(e, t, n || GF);
}
function GF(e, t) {
  return e > t ? 1 : e < t ? -1 : 0;
}
function HF(e, t, n) {
  let s = 0, r = e.length, a = 0, i = false;
  for (; s < r; ) {
    a = s + (r - s >>> 1);
    let o = n(t, e[a]);
    o > 0 ? s = a + 1 : (r = a, i = !o);
  }
  return i ? s : -s - 1;
}
function WS(e, t, n, s, r) {
  return wb(e, t, n, s, r, 0);
}
function US(e, t, n, s, r, a) {
  return wb(e, t, n, s, r, 0, false, a, true);
}
function GS(e, t, n, s, r, a) {
  return wb(e, t, n, s, r, a, true);
}
function wb(e, t, n, s, r, a, i = false, o = false, u = false) {
  let l = [];
  for (let g = 0; g < t.length; g++)
    t[g] > r && l.push({ score: t[g], boxIndex: g, suppressBeginIndex: 0 });
  l.sort(wx);
  let c = a > 0 ? -0.5 / a : 0, p = [], d = [];
  for (; p.length < n && l.length > 0; ) {
    let g = l.pop(), { score: b, boxIndex: y, suppressBeginIndex: v } = g;
    if (b < r)
      break;
    let x = false;
    for (let k = p.length - 1; k >= v; --k) {
      let I = qF(e, y, p[k]);
      if (I >= s) {
        x = true;
        break;
      }
      if (g.score = g.score * jF(s, c, I), g.score <= r)
        break;
    }
    g.suppressBeginIndex = p.length, x || (g.score === b ? (p.push(y), d.push(g.score)) : g.score > r && WF(l, g, wx));
  }
  let h = p.length, f = n - h;
  o && f > 0 && (p.push(...new Array(f).fill(0)), d.push(...new Array(f).fill(0)));
  let m = { selectedIndices: p };
  return i && (m.selectedScores = d), u && (m.validOutputs = h), m;
}
function qF(e, t, n) {
  let s = e.subarray(t * 4, t * 4 + 4), r = e.subarray(n * 4, n * 4 + 4), a = Math.min(s[0], s[2]), i = Math.min(s[1], s[3]), o = Math.max(s[0], s[2]), u = Math.max(s[1], s[3]), l = Math.min(r[0], r[2]), c = Math.min(r[1], r[3]), p = Math.max(r[0], r[2]), d = Math.max(r[1], r[3]), h = (o - a) * (u - i), f = (p - l) * (d - c);
  if (h <= 0 || f <= 0)
    return 0;
  let m = Math.max(a, l), g = Math.max(i, c), b = Math.min(o, p), y = Math.min(u, d), v = Math.max(b - m, 0) * Math.max(y - g, 0);
  return v / (h + f - v);
}
function jF(e, t, n) {
  let s = Math.exp(t * n * n);
  return n <= e ? s : 0;
}
function wx(e, t) {
  return e.score - t.score || e.score === t.score && t.boxIndex - e.boxIndex;
}
async function KF(e, t, n, s = 0.5, r = Number.NEGATIVE_INFINITY) {
  let a = _(e, "boxes", "nonMaxSuppressionAsync"), i = _(t, "scores", "nonMaxSuppressionAsync"), o = eu(a, i, n, s, r);
  n = o.maxOutputSize, s = o.iouThreshold, r = o.scoreThreshold;
  let u = await Promise.all([a.data(), i.data()]), l = u[0], c = u[1], { selectedIndices: p } = WS(l, c, n, s, r);
  return a !== e && a.dispose(), i !== t && i.dispose(), Zt(p, "int32");
}
var XF = KF;
function YF(e, t, n, s = 0.5, r = Number.NEGATIVE_INFINITY, a = 0) {
  let i = _(e, "boxes", "nonMaxSuppression"), o = _(t, "scores", "nonMaxSuppression"), u = eu(i, o, n, s, r, a);
  n = u.maxOutputSize, s = u.iouThreshold, r = u.scoreThreshold, a = u.softNmsSigma;
  let l = { boxes: i, scores: o }, c = { maxOutputSize: n, iouThreshold: s, scoreThreshold: r, softNmsSigma: a }, p = M.runKernel(Eo, l, c);
  return { selectedIndices: p[0], selectedScores: p[1] };
}
var QF = L({ nonMaxSuppressionWithScore_: YF });
async function ZF(e, t, n, s = 0.5, r = Number.NEGATIVE_INFINITY, a = 0) {
  let i = _(e, "boxes", "nonMaxSuppressionAsync"), o = _(t, "scores", "nonMaxSuppressionAsync"), u = eu(i, o, n, s, r, a);
  n = u.maxOutputSize, s = u.iouThreshold, r = u.scoreThreshold, a = u.softNmsSigma;
  let l = await Promise.all([i.data(), o.data()]), c = l[0], p = l[1], { selectedIndices: d, selectedScores: h } = GS(c, p, n, s, r, a);
  return i !== e && i.dispose(), o !== t && o.dispose(), { selectedIndices: Zt(d, "int32"), selectedScores: Zt(h) };
}
var JF = ZF;
function eO(e, t, n, s = 0.5, r = Number.NEGATIVE_INFINITY, a = false) {
  let i = _(e, "boxes", "nonMaxSuppression"), o = _(t, "scores", "nonMaxSuppression"), u = eu(i, o, n, s, r, null), l = u.maxOutputSize, c = u.iouThreshold, p = u.scoreThreshold, d = { boxes: i, scores: o }, h = { maxOutputSize: l, iouThreshold: c, scoreThreshold: p, padToMaxOutputSize: a }, f = M.runKernel(Cl, d, h);
  return { selectedIndices: f[0], validOutputs: f[1] };
}
var tO = L({ nonMaxSuppressionPadded_: eO });
async function nO(e, t, n, s = 0.5, r = Number.NEGATIVE_INFINITY, a = false) {
  let i = _(e, "boxes", "nonMaxSuppressionAsync"), o = _(t, "scores", "nonMaxSuppressionAsync"), u = eu(i, o, n, s, r, null), l = u.maxOutputSize, c = u.iouThreshold, p = u.scoreThreshold, [d, h] = await Promise.all([i.data(), o.data()]), { selectedIndices: f, validOutputs: m } = US(d, h, l, c, p, a);
  return i !== e && i.dispose(), o !== t && o.dispose(), { selectedIndices: Zt(f, "int32"), validOutputs: we(m, "int32") };
}
var sO = nO;
function rO(e, t, n = false, s = false) {
  let r = _(e, "images", "resizeBilinear");
  F(r.rank === 3 || r.rank === 4, () => `Error in resizeBilinear: x must be rank 3 or 4, but got rank ${r.rank}.`), F(t.length === 2, () => `Error in resizeBilinear: new shape must 2D, but got shape ${t}.`), F(s === false || n === false, () => "Error in resizeBilinear: If halfPixelCenters is true, alignCorners must be false.");
  let a = r, i = false;
  r.rank === 3 && (i = true, a = U(r, [1, r.shape[0], r.shape[1], r.shape[2]]));
  let [] = t, o = { images: a }, u = { alignCorners: n, halfPixelCenters: s, size: t }, l = M.runKernel(ri, o, u);
  return i ? U(l, [l.shape[1], l.shape[2], l.shape[3]]) : l;
}
var aO = L({ resizeBilinear_: rO });
function iO(e, t, n = false, s = false) {
  let r = _(e, "images", "resizeNearestNeighbor");
  F(r.rank === 3 || r.rank === 4, () => `Error in resizeNearestNeighbor: x must be rank 3 or 4, but got rank ${r.rank}.`), F(t.length === 2, () => `Error in resizeNearestNeighbor: new shape must 2D, but got shape ${t}.`), F(r.dtype === "float32" || r.dtype === "int32", () => "`images` must have `int32` or `float32` as dtype"), F(s === false || n === false, () => "Error in resizeNearestNeighbor: If halfPixelCenters is true, alignCorners must be false.");
  let a = r, i = false;
  r.rank === 3 && (i = true, a = U(r, [1, r.shape[0], r.shape[1], r.shape[2]]));
  let [] = t, o = { images: a }, u = { alignCorners: n, halfPixelCenters: s, size: t }, l = M.runKernel($l, o, u);
  return i ? U(l, [l.shape[1], l.shape[2], l.shape[3]]) : l;
}
var oO = L({ resizeNearestNeighbor_: iO });
function uO(e, t = "binary", n = false, s = 0.5) {
  let r = _(e, "image", "threshold"), a = 0.2989, i = 0.587, o = 0.114, u = r.shape[0] * r.shape[1], l = V(Zt([s]), 255), c, p, d, h;
  if (F(r.rank === 3, () => `Error in threshold: image must be rank 3,but got rank ${r.rank}.`), F(r.shape[2] === 3 || r.shape[2] === 1, () => `Error in threshold: image color channel must be equal to 3 or 1but got ${r.shape[2]}.`), F(r.dtype === "int32" || r.dtype === "float32", () => `Error in dtype: image dtype must be int32 or float32,but got dtype ${r.dtype}.`), F(t === "otsu" || t === "binary", () => `Method must be binary or otsu, but was ${t}`), r.shape[2] === 3) {
    [c, p, d] = Bn(r, [1, 1, 1], -1);
    let g = V(c, a), b = V(p, i), y = V(d, o);
    h = ie(ie(g, b), y);
  } else
    h = e;
  if (t === "otsu") {
    let g = oS(le(NS(h), "int32"), ms([]), 256);
    l = lO(g, u);
  }
  let f = n ? Jo(h, l) : Un(h, l);
  return le(V(f, 255), "int32");
}
function lO(e, t) {
  let n = Zt([-1]), s = Zt([0]), r = Zt([0]), a, i, o, u, l, c;
  for (let p = 0; p < e.size - 1; p++) {
    a = qe(e, 0, p + 1), i = qe(e, p + 1), l = xe(ve(a), t), c = xe(ve(i), t);
    let d = ve(V(a, el(0, a.size)));
    o = xe(d, ve(a));
    let h = Ll(i.shape, a.size), f = ie(el(0, i.size), h), m = V(i, f);
    u = xe(ve(m), ve(i));
    let g = ge(o, u), b = ge(o, u), y = V(l, c);
    r = V(V(y, g), b);
    let v = Un(r, s);
    s = vn(v, r, s), n = vn(v, Zt([p]), n);
  }
  return n;
}
var cO = L({ threshold_: uO });
function dO(e, t, n = "nearest", s = "constant", r = 0, a) {
  let i = _(e, "image", "transform", "float32"), o = _(t, "transforms", "transform", "float32");
  F(i.rank === 4, () => `Error in transform: image must be rank 4,but got rank ${i.rank}.`), F(o.rank === 2 && (o.shape[0] === i.shape[0] || o.shape[0] === 1) && o.shape[1] === 8, () => "Error in transform: Input transform should be batch x 8 or 1 x 8"), F(a == null || a.length === 2, () => `Error in transform: outputShape must be [height, width] or null, but got ${a}.`);
  let u = { image: i, transforms: o }, l = { interpolation: n, fillMode: s, fillValue: r, outputShape: a };
  return M.runKernel(jo, u, l);
}
var pO = L({ transform_: dO });
function hO(e, t, n) {
  F(t % 1 === 0, () => `bandPart(): numLower must be an integer, got ${t}.`), F(n % 1 === 0, () => `bandPart(): numUpper must be an integer, got ${n}.`);
  let s = _(e, "a", "bandPart");
  F(s.rank >= 2, () => `bandPart(): Rank must be at least 2, got ${s.rank}.`);
  let r = s.shape, [a, i] = s.shape.slice(-2);
  if (!(t <= a))
    throw new Error(`bandPart(): numLower (${t}) must not be greater than the number of rows (${a}).`);
  if (!(n <= i))
    throw new Error(`bandPart(): numUpper (${n}) must not be greater than the number of columns (${i}).`);
  t < 0 && (t = a), n < 0 && (n = i);
  let o = U(el(0, a, 1, "int32"), [-1, 1]), u = el(0, i, 1, "int32"), l = ge(o, u), c = Ds(Jo(l, we(+t, "int32")), Zo(l, we(-n, "int32"))), p = $t([a, i], s.dtype);
  return U(es(Fs(U(s, [-1, a, i])).map((d) => vn(c, d, p))), r);
}
var fO = L({ bandPart_: hO });
function mO(e) {
  let t;
  if (Array.isArray(e)) {
    t = false, F(e != null && e.length > 0, () => "Gram-Schmidt process: input must not be null, undefined, or empty");
    let r = e[0].shape[0];
    for (let a = 1; a < e.length; ++a)
      F(e[a].shape[0] === r, () => `Gram-Schmidt: Non-unique lengths found in the input vectors: (${e[a].shape[0]} vs. ${r})`);
  } else
    t = true, e = Bn(e, e.shape[0], 0).map((r) => mr(r, [0]));
  F(e.length <= e[0].shape[0], () => `Gram-Schmidt: Number of vectors (${e.length}) exceeds number of dimensions (${e[0].shape[0]}).`);
  let n = [], s = e;
  for (let r = 0; r < e.length; ++r)
    n.push(M.tidy(() => {
      let a = s[r];
      if (r > 0)
        for (let i = 0; i < r; ++i) {
          let o = V(ve(V(n[i], a)), n[i]);
          a = ge(a, o);
        }
      return xe(a, sb(a, "euclidean"));
    }));
  return t ? es(n, 0) : n;
}
var gO = L({ gramSchmidt_: mO });
function bO(e, t = false) {
  if (F(e.rank >= 2, () => `qr() requires input tensor to have a rank >= 2, but got rank ${e.rank}`), e.rank === 2)
    return kx(e, t);
  {
    let n = e.shape.slice(0, e.shape.length - 2).reduce((u, l) => u * l), s = Fs(U(e, [n, e.shape[e.shape.length - 2], e.shape[e.shape.length - 1]]), 0), r = [], a = [];
    s.forEach((u) => {
      let [l, c] = kx(u, t);
      r.push(l), a.push(c);
    });
    let i = U(es(r, 0), e.shape), o = U(es(a, 0), e.shape);
    return [i, o];
  }
}
function kx(e, t = false) {
  return M.tidy(() => {
    F(e.shape.length === 2, () => `qr2d() requires a 2D Tensor, but got a ${e.shape.length}D Tensor.`);
    let n = e.shape[0], s = e.shape[1], r = yS(n), a = ur(e), i = Zi([[1]], [1, 1]), o = ur(i), u = n >= s ? s : n;
    for (let l = 0; l < u; ++l) {
      let c = a, p = o, d = r;
      [o, a, r] = M.tidy(() => {
        let h = qe(a, [l, l], [n - l, 1]), f = sb(h), m = qe(a, [l, l], [1, 1]), g = vn(Un(m, 0), Zi([[-1]]), Zi([[1]])), b = ge(m, V(g, f)), y = xe(h, b);
        y.shape[0] === 1 ? o = ur(i) : o = Ot([i, qe(y, [1, 0], [y.shape[0] - 1, y.shape[1]])], 0);
        let v = kt(xe(Ve(g, b), f)), x = qe(a, [l, 0], [n - l, s]), k = V(v, o), I = Ge(o);
        if (l === 0)
          a = ge(x, Ve(k, Ve(I, x)));
        else {
          let E = ge(x, Ve(k, Ve(I, x)));
          a = Ot([qe(a, [0, 0], [l, s]), E], 0);
        }
        let $ = Ge(k), R = qe(r, [0, l], [n, r.shape[1] - l]);
        if (l === 0)
          r = ge(R, Ve(Ve(R, o), $));
        else {
          let E = ge(R, Ve(Ve(R, o), $));
          r = Ot([qe(r, [0, 0], [n, l]), E], 1);
        }
        return [o, a, r];
      }), De([c, p, d]);
    }
    return !t && n > s && (r = qe(r, [0, 0], [n, s]), a = qe(a, [0, 0], [s, s])), [r, a];
  });
}
var yO = L({ qr_: bO });
var vO = ((e) => (e[e.NONE = 0] = "NONE", e[e.MEAN = 1] = "MEAN", e[e.SUM = 2] = "SUM", e[e.SUM_BY_NONZERO_WEIGHTS = 3] = "SUM_BY_NONZERO_WEIGHTS", e))(vO || {});
function xO(e, t, n = 3) {
  let s = _(e, "losses", "computeWeightedLoss"), r = null;
  t != null && (r = _(t, "weights", "computeWeightedLoss"));
  let a = r == null ? s : V(s, r);
  if (n === 0)
    return a;
  if (n === 2)
    return ve(a);
  if (n === 1) {
    if (r == null)
      return It(a);
    {
      let i = s.size / r.size, o = xe(ve(a), ve(r));
      return i > 1 ? xe(o, we(i)) : o;
    }
  }
  if (n === 3) {
    if (r == null)
      return xe(ve(a), we(s.size));
    {
      let i = V(r, Mn(s.shape)), o = le(ve(Ju(i, we(0))), "float32");
      return xe(ve(a), o);
    }
  }
  throw Error(`Unknown reduction: ${n}`);
}
var Ys = L({ computeWeightedLoss_: xO });
function wO(e, t, n, s = 3) {
  let r = _(e, "labels", "absoluteDifference"), a = _(t, "predictions", "absoluteDifference"), i = null;
  n != null && (i = _(n, "weights", "absoluteDifference")), pn(r.shape, a.shape, "Error in absoluteDifference: ");
  let o = Lt(ge(r, a));
  return Ys(o, i, s);
}
var kO = L({ absoluteDifference_: wO });
function SO(e, t, n, s, r = 3) {
  let a = _(e, "labels", "cosineDistance"), i = _(t, "predictions", "cosineDistance"), o = null;
  s != null && (o = _(s, "weights", "cosineDistance")), pn(a.shape, i.shape, "Error in cosineDistance: ");
  let u = we(1), l = ge(u, ve(V(a, i), n, true));
  return Ys(l, o, r);
}
var IO = L({ cosineDistance_: SO });
function CO(e, t, n, s = 3) {
  let r = _(e, "labels", "hingeLoss"), a = _(t, "predictions", "hingeLoss"), i = null;
  n != null && (i = _(n, "weights", "hingeLoss")), pn(r.shape, a.shape, "Error in hingeLoss: ");
  let o = we(1);
  r = ge(V(we(2), r), o);
  let u = Xs(ge(o, V(r, a)));
  return Ys(u, i, s);
}
var NO = L({ hingeLoss_: CO });
function TO(e, t, n, s = 1, r = 3) {
  let a = _(e, "labels", "huberLoss"), i = _(t, "predictions", "huberLoss"), o = null;
  n != null && (o = _(n, "weights", "huberLoss")), pn(a.shape, i.shape, "Error in huberLoss: ");
  let u = we(s), l = Lt(ge(i, a)), c = Ip(l, u), p = ge(l, c), d = ie(V(we(0.5), ct(c)), V(u, p));
  return Ys(d, o, r);
}
var $O = L({ huberLoss_: TO });
function _O(e, t, n, s = 1e-7, r = 3) {
  let a = _(e, "labels", "logLoss"), i = _(t, "predictions", "logLoss"), o = null;
  n != null && (o = _(n, "weights", "logLoss")), pn(a.shape, i.shape, "Error in logLoss: ");
  let u = we(1), l = we(s), c = kt(V(a, Qn(ie(i, l)))), p = V(ge(u, a), Qn(ie(ge(u, i), l))), d = ge(c, p);
  return Ys(d, o, r);
}
var AO = L({ logLoss_: _O });
function EO(e, t, n, s = 3) {
  let r = _(e, "labels", "meanSquaredError"), a = _(t, "predictions", "meanSquaredError"), i = null;
  n != null && (i = _(n, "weights", "meanSquaredError")), pn(r.shape, a.shape, "Error in meanSquaredError: ");
  let o = DS(r, a);
  return Ys(o, i, s);
}
var RO = L({ meanSquaredError_: EO });
function DO(e, t) {
  let n = _(e, "labels", "sigmoidCrossEntropyWithLogits"), s = _(t, "logits", "sigmoidCrossEntropyWithLogits");
  pn(n.shape, s.shape, "Error in sigmoidCrossEntropyWithLogits: ");
  let r = Xs(s), a = V(s, n), i = ib(Yn(kt(Lt(s))));
  return ie(ge(r, a), i);
}
function FO(e, t, n, s = 0, r = 3) {
  let a = _(e, "multiClassLabels", "sigmoidCrossEntropy"), i = _(t, "logits", "sigmoidCrossEntropy"), o = null;
  if (n != null && (o = _(n, "weights", "sigmoidCrossEntropy")), pn(a.shape, i.shape, "Error in sigmoidCrossEntropy: "), s > 0) {
    let l = we(s), c = we(1), p = we(0.5);
    a = ie(V(a, ge(c, l)), V(p, l));
  }
  let u = DO(a, i);
  return Ys(u, o, r);
}
var OO = L({ sigmoidCrossEntropy_: FO });
function PO(e, t, n = -1) {
  if (n === -1 && (n = t.rank - 1), n !== t.rank - 1)
    throw Error(`Softmax cross entropy along a non-last dimension is not yet supported. Labels / logits was rank ${t.rank} and dim was ${n}`);
  return qs((r, a, i) => {
    let u = xD(a, [n], true), l = ge(le(a, "float32"), u);
    i([r, l]);
    let c = kt(V(l, r));
    return { value: ve(c, [n]), gradFunc: (h, f) => {
      let [m, g] = f, b = pa(h.shape, [n]);
      return [V(U(h, b), ge(le(m, "float32"), Yn(g))), V(U(h, b), ge(Yn(g), le(m, "float32")))];
    } };
  })(e, t);
}
function zO(e, t, n, s = 0, r = 3) {
  let a = _(e, "onehotLabels", "softmaxCrossEntropy"), i = _(t, "logits", "softmaxCrossEntropy"), o = null;
  if (n != null && (o = _(n, "weights", "softmaxCrossEntropy")), pn(a.shape, i.shape, "Error in softmaxCrossEntropy: "), s > 0) {
    let l = we(s), c = we(1), p = we(a.shape[1]);
    a = ie(V(a, ge(c, l)), xe(l, p));
  }
  let u = PO(a, i);
  return Ys(u, o, r);
}
var MO = L({ softmaxCrossEntropy_: zO });
function LO(e, t, n, s) {
  let r = _(e, "indices", "sparseFillEmptyRows", "int32"), a = _(t, "values", "sparseFillEmptyRows"), i = _(n, "denseShape", "sparseFillEmptyRows", "int32"), o = _(s, "defaultValue", "sparseFillEmptyRows", a.dtype);
  if (r.rank !== 2)
    throw new Error(`Indices should be Tensor2D but received shape
        ${r.shape}`);
  if (a.rank !== 1)
    throw new Error(`Values should be Tensor1D but received shape ${a.shape}`);
  if (i.rank !== 1)
    throw new Error(`Dense shape should be Tensor1D but received shape ${i.shape}`);
  if (o.rank !== 0)
    throw new Error(`Default value should be a scalar but received shape ${o.shape}`);
  let u = { indices: r, values: a, denseShape: i, defaultValue: o }, l = M.runKernel(cp, u);
  return { outputIndices: l[0], outputValues: l[1], emptyRowIndicator: l[2], reverseIndexMap: l[3] };
}
var BO = L({ sparseFillEmptyRows_: LO });
function VO(e, t, n) {
  let s = _(e, "inputIndices", "sparseReshape", "int32"), r = _(t, "inputShape", "sparseReshape", "int32"), a = _(n, "newShape", "sparseReshape", "int32");
  if (s.rank !== 2)
    throw new Error(`Input indices should be Tensor2D but received shape
        ${s.shape}`);
  if (r.rank !== 1)
    throw new Error(`Input shape should be Tensor1D but received shape ${r.shape}`);
  if (a.rank !== 1)
    throw new Error(`New shape should be Tensor1D but received shape ${a.shape}`);
  let i = { inputIndices: s, inputShape: r, newShape: a }, o = M.runKernel(Rl, i);
  return { outputIndices: o[0], outputShape: o[1] };
}
var WO = L({ sparseReshape_: VO });
function UO(e, t, n) {
  let s = _(e, "data", "sparseSegmentMean"), r = _(t, "indices", "sparseSegmentMean", "int32"), a = _(n, "segmentIds", "sparseSegmentMean", "int32");
  if (s.rank < 1)
    throw new Error("Data should be at least 1 dimensional but received scalar");
  if (r.rank !== 1)
    throw new Error(`Indices should be Tensor1D but received shape
          ${r.shape}`);
  if (a.rank !== 1)
    throw new Error(`Segment ids should be Tensor1D but received shape
          ${a.shape}`);
  let i = { data: s, indices: r, segmentIds: a };
  return M.runKernel(dp, i);
}
var GO = L({ sparseSegmentMean_: UO });
function HO(e, t, n) {
  let s = _(e, "data", "sparseSegmentSum"), r = _(t, "indices", "sparseSegmentSum", "int32"), a = _(n, "segmentIds", "sparseSegmentSum", "int32");
  if (s.rank < 1)
    throw new Error("Data should be at least 1 dimensional but received scalar");
  if (r.rank !== 1)
    throw new Error(`Indices should be Tensor1D but received shape
         ${r.shape}`);
  if (a.rank !== 1)
    throw new Error(`Segment ids should be Tensor1D but received shape
         ${a.shape}`);
  let i = { data: s, indices: r, segmentIds: a };
  return M.runKernel(pp, i);
}
var qO = L({ sparseSegmentSum_: HO });
function jO(e, t, n, s, r, a, i, o) {
  let u = _(e, "data", "stringNGrams", "string");
  if (u.dtype !== "string")
    throw new Error("Data must be of datatype string");
  if (u.shape.length !== 1)
    throw new Error(`Data must be a vector, saw: ${u.shape}`);
  let l = _(t, "dataSplits", "stringNGrams");
  if (l.dtype !== "int32")
    throw new Error("Data splits must be of datatype int32");
  let c = { separator: n, nGramWidths: s, leftPad: r, rightPad: a, padWidth: i, preserveShortSequences: o }, p = { data: u, dataSplits: l }, d = M.runKernel(fp, p, c);
  return { nGrams: d[0], nGramsSplits: d[1] };
}
var KO = L({ stringNGrams_: jO });
function XO(e, t, n = true) {
  let s = _(e, "input", "stringSplit", "string"), r = _(t, "delimiter", "stringSplit", "string");
  if (s.rank !== 1)
    throw new Error(`Input should be Tensor1D but received shape ${s.shape}`);
  if (r.rank !== 0)
    throw new Error(`Delimiter should be a scalar but received shape ${r.shape}`);
  let a = { skipEmpty: n }, i = { input: s, delimiter: r }, o = M.runKernel(Og, i, a);
  return { indices: o[0], values: o[1], shape: o[2] };
}
var YO = L({ stringSplit_: XO });
function QO(e, t) {
  let n = _(e, "input", "stringToHashBucketFast", "string"), s = { numBuckets: t };
  if (t <= 0)
    throw new Error("Number of buckets must be at least 1");
  let r = { input: n };
  return M.runKernel(Pg, r, s);
}
var ZO = L({ stringToHashBucketFast_: QO });
var the = { fft: bb, ifft: Td, rfft: yb, irfft: RS };
var nhe = { hammingWindow: TF, hannWindow: BS, frame: VS, stft: EF };
var jn = { flipLeftRight: OF, grayscaleToRGB: zF, resizeNearestNeighbor: oO, resizeBilinear: aO, rotateWithOffset: LF, cropAndResize: DF, nonMaxSuppression: VF, nonMaxSuppressionAsync: XF, nonMaxSuppressionWithScore: QF, nonMaxSuppressionWithScoreAsync: JF, nonMaxSuppressionPadded: tO, nonMaxSuppressionPaddedAsync: sO, threshold: cO, transform: pO };
var JO = { bandPart: fO, gramSchmidt: gO, qr: yO };
var she = { absoluteDifference: kO, computeWeightedLoss: Ys, cosineDistance: IO, hingeLoss: NO, huberLoss: $O, logLoss: AO, meanSquaredError: RO, sigmoidCrossEntropy: OO, softmaxCrossEntropy: MO };
var Hc = { sparseFillEmptyRows: BO, sparseReshape: WO, sparseSegmentMean: GO, sparseSegmentSum: qO };
var Hf = { stringNGrams: KO, stringSplit: YO, stringToHashBucketFast: ZO };
var _r = class extends Qk {
  minimize(e, t = false, n) {
    let { value: s, grads: r } = this.computeGradients(e, n);
    if (n != null) {
      let a = n.map((i) => ({ name: i.name, tensor: r[i.name] }));
      this.applyGradients(a);
    } else
      this.applyGradients(r);
    return De(r), t ? s : (s.dispose(), null);
  }
  get iterations() {
    return this.iterations_ == null && (this.iterations_ = 0), this.iterations_;
  }
  incrementIterations() {
    this.iterations_ = this.iterations + 1;
  }
  computeGradients(e, t) {
    return hD(e, t);
  }
  dispose() {
    this.iterations_ != null && De(this.iterations_);
  }
  async saveIterations() {
    return this.iterations_ == null && (this.iterations_ = 0), { name: "iter", tensor: we(this.iterations_, "int32") };
  }
  async getWeights() {
    throw new Error("getWeights() is not implemented for this optimizer yet.");
  }
  async setWeights(e) {
    throw new Error(`setWeights() is not implemented for this optimizer class ${this.getClassName()}`);
  }
  async extractIterations(e) {
    return this.iterations_ = (await e[0].tensor.data())[0], e.slice(1);
  }
};
Object.defineProperty(_r, Symbol.hasInstance, { value: (e) => e.minimize != null && e.computeGradients != null && e.applyGradients != null });
var kb = class extends _r {
  constructor(e, t, n = null) {
    super(), this.learningRate = e, this.rho = t, this.epsilon = n, this.accumulatedGrads = [], this.accumulatedUpdates = [], n == null && (this.epsilon = M.backend.epsilon());
  }
  applyGradients(e) {
    (Array.isArray(e) ? e.map((n) => n.name) : Object.keys(e)).forEach((n, s) => {
      let r = M.registeredVariables[n], a = false;
      this.accumulatedGrads[s] == null && (this.accumulatedGrads[s] = { originalName: `${n}/accum_grad`, variable: j(() => je(r).variable(a)) }), this.accumulatedUpdates[s] == null && (this.accumulatedUpdates[s] = { originalName: `${n}/accum_var`, variable: j(() => je(r).variable(a)) });
      let i = Array.isArray(e) ? e[s].tensor : e[n];
      if (i == null)
        return;
      let o = this.accumulatedGrads[s].variable, u = this.accumulatedUpdates[s].variable;
      j(() => {
        let l = ie(V(o, this.rho), V(ct(i), 1 - this.rho)), c = V(xe(dn(ie(u, this.epsilon)), dn(ie(o, this.epsilon))), i), p = ie(V(u, this.rho), V(ct(c), 1 - this.rho));
        o.assign(l), u.assign(p);
        let d = ie(V(c, -this.learningRate), r);
        r.assign(d);
      });
    }), this.incrementIterations();
  }
  dispose() {
    this.accumulatedUpdates != null && (De(this.accumulatedGrads.map((e) => e.variable)), De(this.accumulatedUpdates.map((e) => e.variable)));
  }
  async getWeights() {
    let e = [...this.accumulatedGrads, ...this.accumulatedUpdates];
    return [await this.saveIterations()].concat(e.map((t) => ({ name: t.originalName, tensor: t.variable })));
  }
  async setWeights(e) {
    e = await this.extractIterations(e);
    let t = e.length / 2, n = false;
    this.accumulatedGrads = e.slice(0, t).map((s) => ({ originalName: s.name, variable: s.tensor.variable(n) })), this.accumulatedUpdates = e.slice(t, t * 2).map((s) => ({ originalName: s.name, variable: s.tensor.variable(n) }));
  }
  getConfig() {
    return { learningRate: this.learningRate, rho: this.rho, epsilon: this.epsilon };
  }
  static fromConfig(e, t) {
    return new e(t.learningRate, t.rho, t.epsilon);
  }
};
kb.className = "Adadelta";
Tr(kb);
var Sb = class extends _r {
  constructor(e, t = 0.1) {
    super(), this.learningRate = e, this.initialAccumulatorValue = t, this.accumulatedGrads = [];
  }
  applyGradients(e) {
    (Array.isArray(e) ? e.map((n) => n.name) : Object.keys(e)).forEach((n, s) => {
      let r = M.registeredVariables[n];
      this.accumulatedGrads[s] == null && (this.accumulatedGrads[s] = { originalName: `${n}/accumulator`, variable: j(() => Ll(r.shape, this.initialAccumulatorValue).variable(false)) });
      let a = Array.isArray(e) ? e[s].tensor : e[n];
      if (a == null)
        return;
      let i = this.accumulatedGrads[s].variable;
      j(() => {
        let o = ie(i, ct(a));
        i.assign(o);
        let u = ie(V(xe(a, dn(ie(o, M.backend.epsilon()))), -this.learningRate), r);
        r.assign(u);
      });
    }), this.incrementIterations();
  }
  dispose() {
    this.accumulatedGrads != null && De(this.accumulatedGrads.map((e) => e.variable));
  }
  async getWeights() {
    return [await this.saveIterations()].concat(this.accumulatedGrads.map((e) => ({ name: e.originalName, tensor: e.variable })));
  }
  async setWeights(e) {
    e = await this.extractIterations(e);
    let t = false;
    this.accumulatedGrads = e.map((n) => ({ originalName: n.name, variable: n.tensor.variable(t) }));
  }
  getConfig() {
    return { learningRate: this.learningRate, initialAccumulatorValue: this.initialAccumulatorValue };
  }
  static fromConfig(e, t) {
    return new e(t.learningRate, t.initialAccumulatorValue);
  }
};
Sb.className = "Adagrad";
Tr(Sb);
var Ib = class extends _r {
  constructor(e, t, n, s = null) {
    super(), this.learningRate = e, this.beta1 = t, this.beta2 = n, this.epsilon = s, this.accumulatedFirstMoment = [], this.accumulatedSecondMoment = [], j(() => {
      this.accBeta1 = we(t).variable(), this.accBeta2 = we(n).variable();
    }), s == null && (this.epsilon = M.backend.epsilon());
  }
  applyGradients(e) {
    let t = Array.isArray(e) ? e.map((n) => n.name) : Object.keys(e);
    j(() => {
      let n = ge(1, this.accBeta1), s = ge(1, this.accBeta2);
      t.forEach((r, a) => {
        let i = M.registeredVariables[r], o = false;
        this.accumulatedFirstMoment[a] == null && (this.accumulatedFirstMoment[a] = { originalName: `${r}/m`, variable: j(() => je(i).variable(o)) }), this.accumulatedSecondMoment[a] == null && (this.accumulatedSecondMoment[a] = { originalName: `${r}/v`, variable: j(() => je(i).variable(o)) });
        let u = Array.isArray(e) ? e[a].tensor : e[r];
        if (u == null)
          return;
        let l = this.accumulatedFirstMoment[a].variable, c = this.accumulatedSecondMoment[a].variable, p = ie(V(l, this.beta1), V(u, 1 - this.beta1)), d = ie(V(c, this.beta2), V(ct(u), 1 - this.beta2)), h = xe(p, n), f = xe(d, s);
        l.assign(p), c.assign(d);
        let m = ie(V(xe(h, ie(dn(f), this.epsilon)), -this.learningRate), i);
        i.assign(m);
      }), this.accBeta1.assign(V(this.accBeta1, this.beta1)), this.accBeta2.assign(V(this.accBeta2, this.beta2));
    }), this.incrementIterations();
  }
  dispose() {
    this.accBeta1.dispose(), this.accBeta2.dispose(), this.accumulatedFirstMoment != null && De(this.accumulatedFirstMoment.map((e) => e.variable)), this.accumulatedSecondMoment != null && De(this.accumulatedSecondMoment.map((e) => e.variable));
  }
  async getWeights() {
    let e = [...this.accumulatedFirstMoment, ...this.accumulatedSecondMoment];
    return [await this.saveIterations()].concat(e.map((t) => ({ name: t.originalName, tensor: t.variable })));
  }
  async setWeights(e) {
    e = await this.extractIterations(e), j(() => {
      this.accBeta1.assign(ha(this.beta1, this.iterations_ + 1)), this.accBeta2.assign(ha(this.beta2, this.iterations_ + 1));
    });
    let t = e.length / 2, n = false;
    this.accumulatedFirstMoment = e.slice(0, t).map((s) => ({ originalName: s.name, variable: s.tensor.variable(n) })), this.accumulatedSecondMoment = e.slice(t, t * 2).map((s) => ({ originalName: s.name, variable: s.tensor.variable(n) }));
  }
  getConfig() {
    return { learningRate: this.learningRate, beta1: this.beta1, beta2: this.beta2, epsilon: this.epsilon };
  }
  static fromConfig(e, t) {
    return new e(t.learningRate, t.beta1, t.beta2, t.epsilon);
  }
};
Ib.className = "Adam";
Tr(Ib);
var Cb = class extends _r {
  constructor(e, t, n, s = null, r = 0) {
    super(), this.learningRate = e, this.beta1 = t, this.beta2 = n, this.epsilon = s, this.decay = r, this.accumulatedFirstMoment = [], this.accumulatedWeightedInfNorm = [], j(() => {
      this.iteration = we(0).variable(), this.accBeta1 = we(t).variable();
    }), s == null && (this.epsilon = M.backend.epsilon());
  }
  applyGradients(e) {
    let t = Array.isArray(e) ? e.map((n) => n.name) : Object.keys(e);
    j(() => {
      let n = ge(1, this.accBeta1), s = xe(-this.learningRate, ie(V(this.iteration, this.decay), 1));
      t.forEach((r, a) => {
        let i = M.registeredVariables[r], o = false;
        this.accumulatedFirstMoment[a] == null && (this.accumulatedFirstMoment[a] = { originalName: `${r}/m`, variable: je(i).variable(o) }), this.accumulatedWeightedInfNorm[a] == null && (this.accumulatedWeightedInfNorm[a] = { originalName: `${r}/v`, variable: je(i).variable(o) });
        let u = Array.isArray(e) ? e[a].tensor : e[r];
        if (u == null)
          return;
        let l = this.accumulatedFirstMoment[a].variable, c = this.accumulatedWeightedInfNorm[a].variable, p = ie(V(l, this.beta1), V(u, 1 - this.beta1)), d = V(c, this.beta2), h = Lt(u), f = $r(d, h);
        l.assign(p), c.assign(f);
        let m = ie(V(xe(s, n), xe(p, ie(f, this.epsilon))), i);
        i.assign(m);
      }), this.iteration.assign(ie(this.iteration, 1)), this.accBeta1.assign(V(this.accBeta1, this.beta1));
    }), this.incrementIterations();
  }
  dispose() {
    this.accBeta1.dispose(), this.iteration.dispose(), this.accumulatedFirstMoment != null && De(this.accumulatedFirstMoment.map((e) => e.variable)), this.accumulatedWeightedInfNorm != null && De(this.accumulatedWeightedInfNorm.map((e) => e.variable));
  }
  async getWeights() {
    throw new Error("getWeights() is not implemented for Adamax yet.");
  }
  async setWeights(e) {
    throw new Error("setWeights() is not implemented for Adamax yet.");
  }
  getConfig() {
    return { learningRate: this.learningRate, beta1: this.beta1, beta2: this.beta2, epsilon: this.epsilon, decay: this.decay };
  }
  static fromConfig(e, t) {
    return new e(t.learningRate, t.beta1, t.beta2, t.epsilon, t.decay);
  }
};
Cb.className = "Adamax";
Tr(Cb);
var Ap = class extends _r {
  constructor(e) {
    super(), this.learningRate = e, this.setLearningRate(e);
  }
  applyGradients(e) {
    (Array.isArray(e) ? e.map((n) => n.name) : Object.keys(e)).forEach((n, s) => {
      let r = Array.isArray(e) ? e[s].tensor : e[n];
      if (r == null)
        return;
      let a = M.registeredVariables[n];
      j(() => {
        let i = ie(V(this.c, r), a);
        a.assign(i);
      });
    }), this.incrementIterations();
  }
  setLearningRate(e) {
    this.learningRate = e, this.c != null && this.c.dispose(), this.c = qt(we(-e));
  }
  dispose() {
    this.c.dispose();
  }
  async getWeights() {
    return [await this.saveIterations()];
  }
  async setWeights(e) {
    if (e = await this.extractIterations(e), e.length !== 0)
      throw new Error("SGD optimizer does not have settable weights.");
  }
  getConfig() {
    return { learningRate: this.learningRate };
  }
  static fromConfig(e, t) {
    return new e(t.learningRate);
  }
};
Ap.className = "SGD";
Tr(Ap);
var Nb = class extends Ap {
  constructor(e, t, n = false) {
    super(e), this.learningRate = e, this.momentum = t, this.useNesterov = n, this.accumulations = [], this.m = we(this.momentum);
  }
  applyGradients(e) {
    (Array.isArray(e) ? e.map((n) => n.name) : Object.keys(e)).forEach((n, s) => {
      let r = M.registeredVariables[n];
      this.accumulations[s] == null && (this.accumulations[s] = { originalName: `${n}/momentum`, variable: j(() => je(r).variable(false)) });
      let a = this.accumulations[s].variable, i = Array.isArray(e) ? e[s].tensor : e[n];
      i != null && j(() => {
        let o, u = ie(V(this.m, a), i);
        this.useNesterov ? o = ie(V(this.c, ie(i, V(u, this.m))), r) : o = ie(V(this.c, u), r), a.assign(u), r.assign(o);
      });
    }), this.incrementIterations();
  }
  dispose() {
    this.m.dispose(), this.accumulations != null && De(this.accumulations.map((e) => e.variable));
  }
  setMomentum(e) {
    this.momentum = e;
  }
  async getWeights() {
    return [await this.saveIterations()].concat(this.accumulations.map((e) => ({ name: e.originalName, tensor: e.variable })));
  }
  async setWeights(e) {
    e = await this.extractIterations(e);
    let t = false;
    this.accumulations = e.map((n) => ({ originalName: n.name, variable: n.tensor.variable(t) }));
  }
  getConfig() {
    return { learningRate: this.learningRate, momentum: this.momentum, useNesterov: this.useNesterov };
  }
  static fromConfig(e, t) {
    return new e(t.learningRate, t.momentum, t.useNesterov);
  }
};
Nb.className = "Momentum";
Tr(Nb);
var Tb = class extends _r {
  constructor(e, t = 0.9, n = 0, s = null, r = false) {
    if (super(), this.learningRate = e, this.decay = t, this.momentum = n, this.epsilon = s, this.accumulatedMeanSquares = [], this.accumulatedMoments = [], this.accumulatedMeanGrads = [], this.centered = r, s == null && (this.epsilon = M.backend.epsilon()), e == null)
      throw new Error("learningRate for RMSPropOptimizer must be defined.");
  }
  applyGradients(e) {
    (Array.isArray(e) ? e.map((n) => n.name) : Object.keys(e)).forEach((n, s) => {
      let r = M.registeredVariables[n], a = false;
      this.accumulatedMeanSquares[s] == null && (this.accumulatedMeanSquares[s] = { originalName: `${n}/rms`, variable: j(() => je(r).variable(a)) }), this.accumulatedMoments[s] == null && (this.accumulatedMoments[s] = { originalName: `${n}/momentum`, variable: j(() => je(r).variable(a)) }), this.accumulatedMeanGrads[s] == null && this.centered && (this.accumulatedMeanGrads[s] = { originalName: `${n}/mg`, variable: j(() => je(r).variable(a)) });
      let i = Array.isArray(e) ? e[s].tensor : e[n];
      if (i == null)
        return;
      let o = this.accumulatedMeanSquares[s].variable, u = this.accumulatedMoments[s].variable;
      j(() => {
        let l = ie(V(o, this.decay), V(ct(i), 1 - this.decay));
        if (this.centered) {
          let c = this.accumulatedMeanGrads[s].variable, p = ie(V(c, this.decay), V(i, 1 - this.decay)), d = xe(V(i, this.learningRate), dn(ge(l, ie(ct(p), this.epsilon)))), h = ie(V(u, this.momentum), d);
          o.assign(l), c.assign(p), u.assign(h);
          let f = ge(r, h);
          r.assign(f);
        } else {
          let c = ie(V(o, this.decay), V(ct(i), 1 - this.decay)), p = ie(V(u, this.momentum), xe(V(i, this.learningRate), dn(ie(c, this.epsilon))));
          o.assign(c), u.assign(p);
          let d = ge(r, p);
          r.assign(d);
        }
      });
    }), this.incrementIterations();
  }
  dispose() {
    this.accumulatedMeanSquares != null && De(this.accumulatedMeanSquares.map((e) => e.variable)), this.accumulatedMeanGrads != null && this.centered && De(this.accumulatedMeanGrads.map((e) => e.variable)), this.accumulatedMoments != null && De(this.accumulatedMoments.map((e) => e.variable));
  }
  async getWeights() {
    let e = [...this.accumulatedMeanSquares, ...this.accumulatedMoments];
    return this.centered && e.push(...this.accumulatedMeanGrads), [await this.saveIterations()].concat(e.map((t) => ({ name: t.originalName, tensor: t.variable })));
  }
  async setWeights(e) {
    e = await this.extractIterations(e);
    let t = this.centered ? e.length / 3 : e.length / 2, n = false;
    this.accumulatedMeanSquares = e.slice(0, t).map((s) => ({ originalName: s.name, variable: s.tensor.variable(n) })), this.accumulatedMoments = e.slice(t, t * 2).map((s) => ({ originalName: s.name, variable: s.tensor.variable(n) })), this.centered && (this.accumulatedMeanGrads = e.slice(t * 2, t * 3).map((s) => ({ originalName: s.name, variable: s.tensor.variable(n) })));
  }
  getConfig() {
    return { learningRate: this.learningRate, decay: this.decay, momentum: this.momentum, epsilon: this.epsilon, centered: this.centered };
  }
  static fromConfig(e, t) {
    return new e(t.learningRate, t.decay, t.momentum, t.epsilon, t.centered);
  }
};
Tb.className = "RMSProp";
Tr(Tb);
var Ur = class {
  static sgd(e) {
    return new Ap(e);
  }
  static momentum(e, t, n = false) {
    return new Nb(e, t, n);
  }
  static rmsprop(e, t = 0.9, n = 0, s = null, r = false) {
    return new Tb(e, t, n, s, r);
  }
  static adam(e = 1e-3, t = 0.9, n = 0.999, s = null) {
    return new Ib(e, t, n, s);
  }
  static adadelta(e = 1e-3, t = 0.95, n = null) {
    return new kb(e, t, n);
  }
  static adamax(e = 2e-3, t = 0.9, n = 0.999, s = null, r = 0) {
    return new Cb(e, t, n, s, r);
  }
  static adagrad(e, t = 0.1) {
    return new Sb(e, t);
  }
};
var Li = { sgd: Ur.sgd, momentum: Ur.momentum, adadelta: Ur.adadelta, adagrad: Ur.adagrad, rmsprop: Ur.rmsprop, adamax: Ur.adamax, adam: Ur.adam };
var eP = (() => typeof requestAnimationFrame != "undefined" ? requestAnimationFrame : typeof setImmediate != "undefined" ? setImmediate : (e) => e())();
function HS() {
  return new Promise((e) => eP(() => e()));
}
var C = {};
Ee(C, { ERF_A1: () => dP, ERF_A2: () => pP, ERF_A3: () => hP, ERF_A4: () => fP, ERF_A5: () => mP, ERF_P: () => cP, PARALLELIZE_THRESHOLD: () => $b, SELU_SCALE: () => jS, SELU_SCALEALPHA: () => qS, applyActivation: () => $p, assertAndGetBroadcastShape: () => rt, assertAxesAreInnerMostDims: () => DR, assertParamsConsistent: () => tP, assignToTypedArray: () => wP, axesAreInnerMostDims: () => tb, calculateShapes: () => Vk, checkEinsumDimSizes: () => TP, checkPadOnDimRoundingMode: () => hn, combineLocations: () => fS, complexWithEvenIndex: () => yP, complexWithOddIndex: () => vP, computeConv2DInfo: () => Ml, computeConv3DInfo: () => rS, computeDefaultPad: () => Yg, computeDilation2DInfo: () => yE, computeOptimalWindowSize: () => sP, computeOutAndReduceShapes: () => mS, computeOutShape: () => nP, computePool2DInfo: () => sS, computePool3DInfo: () => vE, convertConv2DDataFormat: () => aS, decodeEinsumEquation: () => CP, eitherStridesOrDilationsAreOne: () => Ps, expandShapeToKeepDim: () => pa, exponent: () => SP, exponents: () => kP, fromStringArrayToUint8: () => KP, fromUint8ToStringArray: () => jP, getAxesPermutation: () => gS, getBroadcastDims: () => Ok, getComplexWithIndex: () => xP, getEinsumComputePath: () => $P, getEinsumPermutation: () => NP, getFusedBiasGradient: () => Tp, getFusedDyActivation: () => Np, getImageCenter: () => rP, getInnerMostAxes: () => FR, getPermuted: () => iP, getReductionAxes: () => At, getReshaped: () => aP, getReshapedPermuted: () => oP, getSliceBeginCoords: () => uP, getSliceSize: () => lP, getSparseFillEmptyRowsIndicesDenseShapeMismatch: () => RP, getSparseFillEmptyRowsNegativeIndexErrorMessage: () => DP, getSparseFillEmptyRowsOutOfRangeIndexErrorMessage: () => FP, getSparseReshapeEmptyTensorZeroOutputDimErrorMessage: () => zP, getSparseReshapeInputOutputMismatchErrorMessage: () => LP, getSparseReshapeInputOutputMultipleErrorMessage: () => MP, getSparseReshapeMultipleNegativeOneOutputDimErrorMessage: () => OP, getSparseReshapeNegativeOutputDimErrorMessage: () => PP, getSparseSegmentReductionIndicesOutOfRangeErrorMessage: () => UP, getSparseSegmentReductionNegativeSegmentIdsErrorMessage: () => BP, getSparseSegmentReductionNonIncreasingSegmentIdsErrorMessage: () => VP, getSparseSegmentReductionSegmentIdOutOfRangeErrorMessage: () => WP, getUndoAxesPermutation: () => nb, isIdentityPermutation: () => _P, log: () => B$, mergeRealAndImagArrays: () => gP, prepareAndValidate: () => Lk, prepareSplitSize: () => EP, segment_util: () => KS, shouldFuse: () => _p, slice_util: () => wt, splitRealAndImagArrays: () => bP, tupleValuesAreOne: () => fr, upcastType: () => cn, validateInput: () => jg, validateUpdateShape: () => qg, warn: () => rr });
function tP(e, t) {
  let n = e[0].length;
  e.forEach((r, a) => {
    F(r.length === n, () => `Error in concat${n}D: rank of tensors[${a}] must be the same as the rank of the rest (${n})`);
  }), F(t >= 0 && t < n, () => `Error in concat${n}D: axis must be between 0 and ${n - 1}.`);
  let s = e[0];
  e.forEach((r, a) => {
    for (let i = 0; i < n; i++)
      F(i === t || r[i] === s[i], () => `Error in concat${n}D: Shape of tensors[${a}] (${r}) does not match the shape of the rest (${s}) along the non-concatenated axis ${a}.`);
  });
}
function nP(e, t) {
  let n = e[0].slice();
  for (let s = 1; s < e.length; s++)
    n[t] += e[s][t];
  return n;
}
var $b = 30;
function sP(e) {
  return e <= $b ? e : bd(e, Math.floor(Math.sqrt(e)));
}
function rP(e, t, n) {
  let s = n * (typeof e == "number" ? e : e[0]), r = t * (typeof e == "number" ? e : e[1]);
  return [s, r];
}
function aP(e, t, n, s = true) {
  let r = [];
  if (s)
    r = r.concat(t.slice(0)), r.push(e[0] / n), r = r.concat(e.slice(1));
  else {
    r = r.concat(e[0]);
    let a = t.length;
    for (let i = 0; i < a; ++i)
      r = r.concat([e[i + 1] / t[i], t[i]]);
    r = r.concat(e.slice(a + 1));
  }
  return r;
}
function iP(e, t, n = true) {
  let s = [];
  if (n) {
    s.push(t);
    for (let r = t + 1; r < e; ++r)
      r <= 2 * t ? (s.push(r), s.push(r - (t + 1))) : s.push(r);
  } else {
    let r = [], a = [];
    for (let i = 1; i < e; ++i)
      i >= t * 2 + 1 || i % 2 === 1 ? a.push(i) : r.push(i);
    s.push(...r), s.push(0), s.push(...a);
  }
  return s;
}
function oP(e, t, n, s = true) {
  let r = [];
  s ? r.push(e[0] / n) : r.push(e[0] * n);
  for (let a = 1; a < e.length; ++a)
    a <= t.length ? s ? r.push(t[a - 1] * e[a]) : r.push(e[a] / t[a - 1]) : r.push(e[a]);
  return r;
}
function uP(e, t) {
  let n = [0];
  for (let s = 0; s < t; ++s)
    n.push(e[s][0]);
  return n;
}
function lP(e, t, n) {
  let s = e.slice(0, 1);
  for (let r = 0; r < n; ++r)
    s.push(e[r + 1] - t[r][0] - t[r][1]);
  return s;
}
var qS = 1.7580993408473768;
var jS = 1.0507009873554805;
var cP = 0.3275911;
var dP = 0.254829592;
var pP = -0.284496736;
var hP = 1.421413741;
var fP = -1.453152027;
var mP = 1.061405429;
function gP(e, t) {
  if (e.length !== t.length)
    throw new Error(`Cannot merge real and imag arrays of different lengths. real:${e.length}, imag: ${t.length}.`);
  let n = new Float32Array(e.length * 2);
  for (let s = 0; s < n.length; s += 2)
    n[s] = e[s / 2], n[s + 1] = t[s / 2];
  return n;
}
function bP(e) {
  let t = new Float32Array(e.length / 2), n = new Float32Array(e.length / 2);
  for (let s = 0; s < e.length; s += 2)
    t[s / 2] = e[s], n[s / 2] = e[s + 1];
  return { real: t, imag: n };
}
function yP(e) {
  let t = Math.ceil(e.length / 4), n = new Float32Array(t), s = new Float32Array(t);
  for (let r = 0; r < e.length; r += 4)
    n[Math.floor(r / 4)] = e[r], s[Math.floor(r / 4)] = e[r + 1];
  return { real: n, imag: s };
}
function vP(e) {
  let t = Math.floor(e.length / 4), n = new Float32Array(t), s = new Float32Array(t);
  for (let r = 2; r < e.length; r += 4)
    n[Math.floor(r / 4)] = e[r], s[Math.floor(r / 4)] = e[r + 1];
  return { real: n, imag: s };
}
function xP(e, t) {
  let n = e[t * 2], s = e[t * 2 + 1];
  return { real: n, imag: s };
}
function wP(e, t, n, s) {
  e[s * 2] = t, e[s * 2 + 1] = n;
}
function kP(e, t) {
  let n = new Float32Array(e / 2), s = new Float32Array(e / 2);
  for (let r = 0; r < Math.ceil(e / 2); r++) {
    let a = (t ? 2 : -2) * Math.PI * (r / e);
    n[r] = Math.cos(a), s[r] = Math.sin(a);
  }
  return { real: n, imag: s };
}
function SP(e, t, n) {
  let s = (n ? 2 : -2) * Math.PI * (e / t), r = Math.cos(s), a = Math.sin(s);
  return { real: r, imag: a };
}
var qf = "->";
var IP = /->/g;
var Sx = ",";
var Ix = "...";
function CP(e, t) {
  e = e.replace(/\s/g, "");
  let n = (e.length - e.replace(IP, "").length) / qf.length;
  if (n < 1)
    throw new Error("Equations without an arrow are not supported.");
  if (n > 1)
    throw new Error(`Equation must contain exactly one arrow ("${qf}").`);
  let [s, r] = e.split(qf);
  F(s.indexOf(Ix) === -1, () => `The ellipsis notation ("${Ix}") is not supported yet.`);
  let a = s.split(Sx), i = a.length;
  if (t !== i)
    throw new Error(`Expected ${i} input tensors, received ${t}`);
  if (i > 2)
    throw new Error("Support for more than 2 input tensors is not implemented yet.");
  let o = [];
  for (let d = 0; d < r.length; ++d) {
    let h = r[d];
    if (!a.some((f) => f.indexOf(h) !== -1))
      throw new Error(`Output subscripts contain the label ${h} not present in the input subscripts.`);
    o.indexOf(h) === -1 && o.push(h);
  }
  for (let d = 0; d < s.length; ++d) {
    let h = s[d];
    o.indexOf(h) === -1 && h !== Sx && o.push(h);
  }
  let u = new Array(a.length);
  for (let d = 0; d < i; ++d) {
    if (new Set(a[d].split("")).size !== a[d].length)
      throw new Error(`Found duplicate axes in input component ${a[d]}. Support for duplicate axes in input is not implemented yet.`);
    u[d] = [];
    for (let h = 0; h < a[d].length; ++h)
      u[d].push(o.indexOf(a[d][h]));
  }
  let l = o.length, c = r.length, p = [];
  for (let d = c; d < l; ++d)
    p.push(d);
  return { allDims: o, summedDims: p, idDims: u };
}
function NP(e, t) {
  let n = new Array(e);
  n.fill(-1);
  for (let r = 0; r < t.length; ++r)
    n[t[r]] = r;
  let s = [];
  for (let r = 0; r < e; ++r)
    n[r] === -1 && s.push(r);
  return n = n.filter((r) => r !== -1), { permutationIndices: n, expandDims: s };
}
function TP(e, t, n) {
  let s = new Array(e);
  for (let r = 0; r < n.length; ++r) {
    let a = n[r].shape;
    for (let i = 0; i < t[r].length; ++i)
      s[t[r][i]] === void 0 ? s[t[r][i]] = a[i] : F(s[t[r][i]] === a[i], () => `Expected dimension ${s[t[r][i]]} at axis ${i} of input shaped ${JSON.stringify(a)}, but got dimension ${a[i]}`);
  }
}
function $P(e, t) {
  let n = e, s = [], r = 0;
  e.length === 0 && n.push(-1), r = e.length + 1;
  for (let i = 0; i < r; ++i)
    s.push([]);
  let a = [];
  for (let i = 0; i < n.length; ++i) {
    let o = n[i], u = AP(t, o);
    for (let l of u)
      a.indexOf(l) === -1 && (s[i].push(l), a.push(l));
  }
  return { path: n, steps: s };
}
function _P(e) {
  return e.every((t, n) => t === n);
}
function AP(e, t) {
  let n = [];
  for (let s = 0; s < e.length; ++s)
    (e[s].length === 0 || e[s].indexOf(t) !== -1 || t === -1) && n.push(s);
  return n;
}
function EP(e, t, n = 0) {
  let s = [];
  if (typeof t == "number")
    F(e.shape[n] % t === 0, () => "Number of splits must evenly divide the axis."), s = new Array(t).fill(e.shape[n] / t);
  else {
    let r = t.reduce((i, o) => (o === -1 && (i += 1), i), 0);
    F(r <= 1, () => "There should be only one negative value in split array.");
    let a = t.indexOf(-1);
    if (a !== -1) {
      let i = t.reduce((o, u) => u > 0 ? o + u : o);
      t[a] = e.shape[n] - i;
    }
    F(e.shape[n] === t.reduce((i, o) => i + o), () => "The sum of sizes must match the size of the axis dimension."), s = t;
  }
  return s;
}
function RP(e) {
  return `Received SparseTensor with denseShape[0] = 0 but
  indices.shape[0] = ${e}`;
}
function DP(e, t) {
  return `indices(${e}, 0) is invalid: ${t} < 0`;
}
function FP(e, t, n) {
  return `indices(${e}, 0) is invalid: ${t} >= ${n}`;
}
function OP(e, t) {
  return `only one output dimension may be -1, not both ${e} and ${t}`;
}
function PP(e, t) {
  return `size ${e} must be non-negative, not ${t}`;
}
function zP() {
  return "reshape cannot infer the missing input size for an empty tensor unless all specified input sizes are non-zero";
}
function MP(e, t) {
  let n = dt(e), s = dt(t);
  return `Input to reshape is a SparseTensor with ${n}
  dense values, but the requested shape requires a multiple of ${s}. inputShape=${e} outputShape= ${t}`;
}
function LP(e, t) {
  let n = dt(e), s = dt(t);
  return `Input to reshape is a tensor with ${n} dense values, but the requested shape has ${s}. inputShape=${e} outputShape=${t}`;
}
function BP() {
  return "segment ids must be >= 0";
}
function VP() {
  return "segment ids are not increasing";
}
function WP(e, t) {
  return `Segment id ${e} out of range [0, ${t}), possibly because segmentIds input is not sorted.`;
}
function UP(e, t, n) {
  return `Bad: indices[${e}] == ${t} out of range [0, ${n})`;
}
var KS = {};
Ee(KS, { collectGatherOpShapeInfo: () => qP, computeOutShape: () => HP, segOpComputeOptimalWindowSize: () => GP });
function GP(e, t) {
  let n = false, s;
  for (e <= $b ? (s = e, n = true) : s = bd(e, Math.floor(Math.sqrt(e))); !n; )
    s > t || s === e ? n = true : s = bd(e, s + 1);
  return s;
}
function HP(e, t, n) {
  let s = [], r = e.length;
  for (let a = 0; a < r; a++)
    a !== t ? s.push(e[a]) : s.push(n);
  return s;
}
function qP(e, t, n, s) {
  let r = t.shape.length, a = e.shape.length;
  if (s !== 0 && (s < -r || s > r))
    throw new Error(`Expect batchDims in the range of [-${r}, ${r}], but got ${s}`);
  if (s < 0 && (s += r), s > a)
    throw new Error(`batchDims (${s}) must be less than rank(x) (
    ${a}).`);
  if (n < s)
    throw new Error(`batchDims (${s}) must be less than or equal to axis (${n}).`);
  for (let p = 0; p < s; ++p)
    if (e.shape[p] !== t.shape[p])
      throw new Error(`x.shape[${p}]: ${e.shape[p]} should be equal to indices.shape[${p}]: ${t.shape[p]}.`);
  let i = e.shape[n], o = [], u = 1, l = 1, c = 1;
  for (let p = 0; p < s; ++p)
    o.push(e.shape[p]), u *= e.shape[p];
  for (let p = s; p < n; p++)
    o.push(e.shape[p]), l *= e.shape[p];
  for (let p = s; p < r; p++)
    o.push(t.shape[p]);
  for (let p = n + 1; p < a; p++)
    o.push(e.shape[p]), c *= e.shape[p];
  return { batchSize: u, sliceSize: c, outerSize: l, dimSize: i, outputShape: o };
}
function jP(e) {
  try {
    return e.map((t) => vd(t));
  } catch (t) {
    throw new Error(`Failed to decode encoded string bytes into utf-8, error: ${t}`);
  }
}
function KP(e) {
  return e.map((t) => Pl(t));
}
var ws = {};
Ee(ws, { nonMaxSuppressionV3Impl: () => WS, nonMaxSuppressionV4Impl: () => US, nonMaxSuppressionV5Impl: () => GS, whereImpl: () => FS });
var Bs = class extends Error {
  constructor(e) {
    super(e), Object.setPrototypeOf(this, Bs.prototype);
  }
};
var fs = class extends Error {
  constructor(e) {
    super(e), Object.setPrototypeOf(this, fs.prototype);
  }
};
var G = class extends Error {
  constructor(e) {
    super(e), Object.setPrototypeOf(this, G.prototype);
  }
};
var Fe = class extends Error {
  constructor(e) {
    super(e), Object.setPrototypeOf(this, Fe.prototype);
  }
};
var XS = class extends Error {
  constructor(e) {
    super(e), Object.setPrototypeOf(this, XS.prototype);
  }
};
var YS = class {
  constructor(e) {
    this.maxEntries = e || 100, this.cache = /* @__PURE__ */ new Map();
  }
  get(e) {
    let t;
    return this.cache.has(e) && (t = this.cache.get(e), this.cache.delete(e), this.cache.set(e, t)), t;
  }
  put(e, t) {
    if (this.cache.has(e))
      this.cache.delete(e);
    else if (this.cache.size >= this.maxEntries) {
      let n = this.cache.keys().next().value;
      this.cache.delete(n);
    }
    this.cache.set(e, t);
  }
  getMaxEntries() {
    return this.maxEntries;
  }
  setMaxEntries(e) {
    if (e < 0)
      throw new Error(`The maxEntries of LRU caches must be at least 0, but got ${e}.`);
    if (this.maxEntries > e)
      for (let t = 0; t < this.maxEntries - e; t++) {
        let n = this.cache.keys().next().value;
        this.cache.delete(n);
      }
    this.maxEntries = e;
  }
};
function ma(e, t) {
  if (Array.isArray(e)) {
    let n = [];
    for (let s = 0; s < t; s++)
      n = n.concat(e);
    return n;
  } else {
    let n = new Array(t);
    return n.fill(e), n;
  }
}
function Cs(e, t) {
  if (!e)
    throw new XS(t);
}
function Cx(e, t) {
  let n = 0;
  for (let s of e)
    s === t && n++;
  return n;
}
function bn(e) {
  return e.length === 1 ? e[0] : e;
}
function ht(e) {
  return Array.isArray(e) ? e : [e];
}
function Vs(e) {
  let n = e.replace(/(.)([A-Z][a-z0-9]+)/g, "$1_$2").replace(/([a-z])([A-Z])/g, "$1_$2").toLowerCase();
  return n[0] !== "_" ? n : "private" + n;
}
function Xr(e) {
  return e.length <= 1 || e.indexOf("_") === -1 ? e : e.replace(/[_]+(\w|$)/g, (t, n) => n.toUpperCase());
}
var Hn = {};
function _b(e) {
  if (e == null)
    return null;
  let t = {};
  return t.className = e.getClassName(), t.config = e.getConfig(), t;
}
function km(e) {
  if (!(e == null || typeof e != "object"))
    if (Array.isArray(e))
      e.forEach((t) => km(t));
    else {
      let t = Object.keys(e);
      for (let n of t) {
        let s = e[n];
        s != null && typeof s == "object" && (!Array.isArray(s) && s.type === "ndarray" && typeof s.value == "number" ? e[n] = s.value : km(s));
      }
    }
}
function Wl(e, t = {}, n = {}, s = "object", r = false) {
  if (typeof e == "string") {
    let a = e, i;
    if (a in n)
      i = n[a];
    else if (a in Hn)
      i = Hn[a];
    else if (i = t[a], i == null)
      throw new G(`Unknown ${s}: ${e}. This may be due to one of the following reasons:
1. The ${s} is defined in Python, in which case it needs to be ported to TensorFlow.js or your JavaScript code.
2. The custom ${s} is defined in JavaScript, but is not registered properly with tf.serialization.registerClass().`);
    return i;
  } else {
    let a = e;
    if (a.className == null || a.config == null)
      throw new G(`${s}: Improper config format: ${JSON.stringify(a)}.
'className' and 'config' must set.`);
    let i = a.className, o, u;
    if (i in n ? [o, u] = n[i] : i in Hn ? [o, u] = Hn.className : i in t && ([o, u] = t[i]), o == null)
      throw new G(`Unknown ${s}: ${i}. This may be due to one of the following reasons:
1. The ${s} is defined in Python, in which case it needs to be ported to TensorFlow.js or your JavaScript code.
2. The custom ${s} is defined in JavaScript, but is not registered properly with tf.serialization.registerClass().`);
    if (u != null) {
      let l = {};
      for (let h of Object.keys(Hn))
        l[h] = Hn[h];
      for (let h of Object.keys(n))
        l[h] = n[h];
      let c = a.config;
      c.customObjects = l;
      let p = { ...Hn };
      for (let h of Object.keys(n))
        Hn[h] = n[h];
      km(a.config);
      let d = u(o, a.config, n, r);
      return Hn = { ...p }, d;
    } else {
      let l = { ...Hn };
      for (let p of Object.keys(n))
        Hn[p] = n[p];
      let c = new o(a.config);
      return Hn = { ...l }, c;
    }
  }
}
function XP(e, t) {
  return e < t ? -1 : e > t ? 1 : 0;
}
function qc(e, t) {
  return -1 * XP(e, t);
}
function lr(e) {
  if (e == null)
    return e;
  let t = [];
  for (let n of e)
    t.indexOf(n) === -1 && t.push(n);
  return t;
}
function YP(e) {
  if (e == null)
    throw new G(`Invalid value in obj: ${JSON.stringify(e)}`);
  for (let t in e)
    if (e.hasOwnProperty(t))
      return false;
  return true;
}
function yi(e, t, n) {
  if (n != null && e.indexOf(n) < 0)
    throw new G(`${n} is not a valid ${t}.  Valid values are ${e} or null/undefined.`);
}
function Ab(e, t, n = 0, s = 1 / 0) {
  return Cs(n >= 0), Cs(s >= n), Array.isArray(e) && e.length >= n && e.length <= s && e.every((r) => typeof r === t);
}
function Vt(e, t) {
  Array.isArray(e) ? (w.assert(e.length > 0, () => `${t} is unexpectedly an empty array.`), e.forEach((n, s) => Vt(n, `element ${s + 1} of ${t}`))) : w.assert(Number.isInteger(e) && e > 0, () => `Expected ${t} to be a positive integer, but got ${QS(e)}.`);
}
function QS(e) {
  return e === null ? "null" : Array.isArray(e) ? "[" + e.map((t) => QS(t)).join(",") + "]" : typeof e == "string" ? `"${e}"` : `${e}`;
}
function QP(e, t, n) {
  let s = n != null ? n() : w.now(), r;
  return (...i) => {
    let o = n != null ? n() : w.now();
    return o - s < t || (s = o, r = e(...i)), r;
  };
}
function ZS(e) {
  return e === "relu" ? "relu" : e === "linear" ? "linear" : e === "elu" ? "elu" : null;
}
var ZP = 0;
function JS() {
  return ZP++;
}
var jc = {};
function Ep(e = "") {
  return e in jc || (jc[e] = 0), jc[e] += 1, e + jc[e].toString();
}
var JP = ["channelsFirst", "channelsLast"];
var ez = ["nearest", "bilinear"];
var tz = ["valid", "same", "causal"];
var nz = ["max", "avg"];
var sz = ["sum", "mul", "concat", "ave"];
var Bi = /* @__PURE__ */ new Map();
function Ct(e) {
  yi(JP, "DataFormat", e);
}
function rz(e) {
  yi(ez, "InterpolationFormat", e);
}
function Gn(e) {
  yi(tz, "PaddingMode", e);
}
function eI(e) {
  yi(nz, "PoolMode", e);
}
var Wu = [];
var Nx = "/";
function ta(e, t) {
  Wu.push(e);
  try {
    let n = t();
    return Wu.pop(), n;
  } catch (n) {
    throw Wu.pop(), n;
  }
}
function az() {
  return Wu.length === 0 ? "" : Wu.join(Nx) + Nx;
}
function tI(e) {
  if (!sI(e))
    throw new Error("Not a valid tensor name: '" + e + "'");
  return az() + e;
}
function nI(e) {
  if (!sI(e))
    throw new Error("Not a valid tensor name: '" + e + "'");
  Bi.has(e) || Bi.set(e, 0);
  let t = Bi.get(e);
  if (Bi.set(e, Bi.get(e) + 1), t > 0) {
    let n = `${e}_${t}`;
    return Bi.set(n, 1), n;
  } else
    return e;
}
var iz = new RegExp(/^[A-Za-z0-9][-A-Za-z0-9\._\/]*$/);
function sI(e) {
  return !!e.match(iz);
}
function oz(e) {
  return e === parseInt(e.toString(), 10);
}
function cr(e, t, n) {
  t == null && (t = 0), n == null && (n = e.length);
  let s = 1;
  for (let r = t; r < n; ++r)
    s *= e[r];
  return s;
}
function no(e) {
  if (e.length === 0)
    return Number.NaN;
  let t = Number.POSITIVE_INFINITY;
  for (let n = 0; n < e.length; n++) {
    let s = e[n];
    s < t && (t = s);
  }
  return t;
}
function gr(e) {
  if (e.length === 0)
    return Number.NaN;
  let t = Number.NEGATIVE_INFINITY;
  for (let n = 0; n < e.length; n++) {
    let s = e[n];
    s > t && (t = s);
  }
  return t;
}
function ys(e, t) {
  if (t < e)
    throw new G(`end (${t}) < begin (${e}) is forbidden.`);
  let n = [];
  for (let s = e; s < t; ++s)
    n.push(s);
  return n;
}
var jf;
function Dt() {
  return jf == null && (jf = HA().epsilon()), jf;
}
function vs() {
  return "channelsLast";
}
function Rp(e, t) {
  return le(e, t);
}
function Ul(e, t = -1) {
  let n = e.shape.slice();
  return t < 0 && (t = n.length + t + 1), n.splice(t, 0, 1), U(e, n);
}
function uz(e, t) {
  return j(() => {
    if (e.shape.length !== 2)
      throw new G(`repeat() expects a rank-2 tensor, but received a rank-${e.shape.length} tensor.`);
    let n = Ul(e, 1);
    return Sm(n, [1, t, 1]);
  });
}
function lz(e) {
  let t = [cr(e.shape)];
  return U(e, t);
}
function cz(e) {
  if (e.rank <= 1)
    throw new G(`batchFlatten requires a minimum rank of 2. Got rank: ${e.rank}.`);
  let t = [e.shape[0], cr(e.shape, 1)];
  return U(e, t);
}
function na(e, t, n) {
  return j(() => {
    switch (e.rank) {
      case 1:
        return fb(e, t, n);
      case 2:
        return ES(e, [t, 0], [n, e.shape[1]]);
      case 3:
        return mb(e, [t, 0, 0], [n, e.shape[1], e.shape[2]]);
      case 4:
        return Nd(e, [t, 0, 0, 0], [n, e.shape[1], e.shape[2], e.shape[3]]);
      case 5:
        return qe(e, [t, 0, 0, 0, 0], [n, e.shape[1], e.shape[2], e.shape[3], e.shape[4]]);
      case 6:
        return qe(e, [t, 0, 0, 0, 0, 0], [n, e.shape[1], e.shape[2], e.shape[3], e.shape[4], e.shape[5]]);
      default:
        throw new G(`sliceAlongFirstAxis() received an unsupported tensor rank: ${e.rank}`);
    }
  });
}
function Kf(e, t, n) {
  return j(() => {
    switch (e.rank) {
      case 1:
        return fb(e, t, n);
      case 2:
        return ES(e, [0, t], [e.shape[0], n]);
      case 3:
        return mb(e, [0, 0, t], [e.shape[0], e.shape[1], n]);
      case 4:
        return Nd(e, [0, 0, 0, t], [e.shape[0], e.shape[1], e.shape[2], n]);
      default:
        throw new G(`sliceAlongLastAxis() received an unsupported tensor rank: ${e.rank}`);
    }
  });
}
function Kc(e, t, n, s) {
  return j(() => {
    switch (e.rank) {
      case 1:
        return fb(e, t, n);
      case 2:
        switch (s) {
          case 1:
            return na(e, t, n);
          case 2:
            return Kf(e, t, n);
          default:
            throw new G(`The axis is not within the rank of the tensor ${s}`);
        }
      case 3:
        switch (s) {
          case 1:
            return na(e, t, n);
          case 2:
            return mb(e, [0, t, 0], [e.shape[0], n, e.shape[2]]);
          case 3:
            return Kf(e, t, n);
          default:
            throw new G(`The axis is not within the rank of the tensor ${s}`);
        }
      case 4:
        switch (s) {
          case 1:
            return na(e, t, n);
          case 2:
            return Nd(e, [0, t, 0, 0], [e.shape[0], n, e.shape[2], e.shape[3]]);
          case 3:
            return Nd(e, [0, 0, t, 0], [e.shape[0], e.shape[1], n, e.shape[3]]);
          case 4:
            return Kf(e, t, n);
          default:
            throw new G(`The axis is not within the rank of the tensor ${s}`);
        }
      default:
        throw new G(`sliceAlongLastAxis() received an unsupported tensor rank: ${e.rank}`);
    }
  });
}
function Eb(e, t = -1) {
  let n;
  return t < 0 && (n = e[0].rank, n !== 0 ? t = n : t = 0), t === e[0].rank && (t = -1), Ot(e, t);
}
function Tx(e, t) {
  switch (e.rank) {
    case 1:
      return XE([e, t]);
    case 2:
      return QE([e, t], 0);
    case 3:
      return JE([e, t], 0);
    case 4:
      return tR([e, t], 0);
    default:
      throw new G(`concatAlongFirstAxis() received an unsupported tensor rank: ${e.rank}`);
  }
}
function Sm(e, t) {
  if (Array.isArray(t) || (t = [t]), e.rank !== t.length)
    throw new G(`The length of input n (${t.length}) does not match the number of dimensions in input x (${e.rank})`);
  return hs(e, t);
}
function Dp(e, t = 0, n = 1, s, r) {
  return o3(e, t, n, s, r);
}
function Es(e, t, n, s) {
  if (e.rank < 2 || t.rank < 2)
    throw new Fe(`dot requires both inputs to be rank >= 2 but got x shape = ${e.shape} and y shape = ${t.shape}`);
  if (t.rank >= 3) {
    let r = e.shape.slice(-1)[0], a = t.shape.slice(-2)[0];
    if (r !== a)
      throw new Fe(`If rank y >= 3, then the second last dim of y must equal the last dim of x but got x shape = ${e.shape} and  y shape = ${t.shape}`);
  }
  if (e.rank === 2 && t.rank === 2)
    return fa.matMul({ a: e, b: t, transposeA: false, transposeB: false, bias: s ? Im(e.rank, s, vs()) : null, activation: n });
  {
    let r = e.shape.slice(), a = r.pop();
    e = U(e, [-1, a]);
    let i = t.shape.slice(), o = i.pop(), u = i.pop(), l = [...i, o], c = Array.from({ length: t.rank }, (f, m) => m === 0 ? t.rank - 2 : m <= t.rank - 2 ? m - 1 : m);
    t = U(Ge(t, c), [u, -1]);
    let p = [...r, ...l], d = false, h = false;
    return U(fa.matMul({ a: e, b: t, transposeA: d, transposeB: h, bias: s ? Im(e.rank, s, vs()) : null, activation: n }), p);
  }
}
function rI(e, t, n) {
  return j(() => (Array.isArray(t) ? t = Zt(t, "int32") : t = le(t, "int32"), Zu(e, t, n)));
}
function Gl(e) {
  return V(e, e);
}
function Im(e, t, n) {
  let s = t.shape;
  if (t.rank !== 1 && t.rank !== e)
    throw new G(`Unexpected bias dimensions: ${t.rank}; expected it to be 1 or ${e}`);
  if (e === 5) {
    if (n === "channelsFirst")
      return s.length === 1 ? U(t, [1, s[0], 1, 1, 1]) : U(t, [1, s[3], s[0], s[1], s[2]]);
    if (n === "channelsLast")
      return s.length === 1 ? U(t, [1, 1, 1, 1, s[0]]) : U(t, [1].concat(s));
  } else if (e === 4) {
    if (n === "channelsFirst")
      return s.length === 1 ? U(t, [1, s[0], 1, 1]) : U(t, [1, s[2], s[0], s[1]]);
    if (n === "channelsLast")
      return s.length === 1 ? U(t, [1, 1, 1, s[0]]) : U(t, [1].concat(s));
  } else if (e === 3) {
    if (n === "channelsFirst")
      return s.length === 1 ? U(t, [1, s[0], 1]) : U(t, [1, s[1], s[0]]);
    if (n === "channelsLast")
      return s.length === 1 ? U(t, [1, 1, s[0]]) : U(t, [1].concat(s));
  } else if (e < 3)
    return t;
  throw new G(`Unsupported input rank by biasAdd: ${t.rank}`);
}
function ks(e, t, n) {
  return j(() => (n == null && (n = vs()), Ct(n), ie(e, Im(e.rank, t, n))));
}
function dz(e, t = 1) {
  if (t !== 1)
    throw new Fe(`Support for alpha values other than 1 (${t}) is not implemented yet.`);
  return wp(e);
}
function pz(e) {
  return j(() => xe(e, ie(Lt(e), 1)));
}
function aI(e, t, n, s) {
  return j(() => fF(e, t, n, s));
}
function hz(e) {
  return j(() => {
    let t = ie(0.5, V(0.2, e));
    return Vn(t, 0, 1);
  });
}
function Hl(e, t, n = false) {
  return n ? e() : t();
}
var fz = ["fanIn", "fanOut", "fanAvg"];
var mz = ["normal", "uniform", "truncatedNormal"];
function gz(e) {
  yi(fz, "FanMode", e);
}
function bz(e) {
  yi(mz, "Distribution", e);
}
var ns = class extends re.Serializable {
  fromConfigUsesCustomObjects() {
    return false;
  }
  getConfig() {
    return {};
  }
};
var Rb = class extends ns {
  apply(e, t) {
    return $t(e, t);
  }
};
Rb.className = "Zeros";
re.registerClass(Rb);
var Fp = class extends ns {
  apply(e, t) {
    return Mn(e, t);
  }
};
Fp.className = "Ones";
re.registerClass(Fp);
var Db = class extends ns {
  constructor(e) {
    if (super(), typeof e != "object")
      throw new G(`Expected argument of type ConstantConfig but got ${e}`);
    if (e.value === void 0)
      throw new G(`config must have value set but got ${e}`);
    this.value = e.value;
  }
  apply(e, t) {
    return j(() => V(we(this.value), Mn(e, t)));
  }
  getConfig() {
    return { value: this.value };
  }
};
Db.className = "Constant";
re.registerClass(Db);
var Fb = class extends ns {
  constructor(e) {
    super(), this.DEFAULT_MINVAL = -0.05, this.DEFAULT_MAXVAL = 0.05, this.minval = e.minval || this.DEFAULT_MINVAL, this.maxval = e.maxval || this.DEFAULT_MAXVAL, this.seed = e.seed;
  }
  apply(e, t) {
    return Vl(e, this.minval, this.maxval, t);
  }
  getConfig() {
    return { minval: this.minval, maxval: this.maxval, seed: this.seed };
  }
};
Fb.className = "RandomUniform";
re.registerClass(Fb);
var Ob = class extends ns {
  constructor(e) {
    super(), this.DEFAULT_MEAN = 0, this.DEFAULT_STDDEV = 0.05, this.mean = e.mean || this.DEFAULT_MEAN, this.stddev = e.stddev || this.DEFAULT_STDDEV, this.seed = e.seed;
  }
  apply(e, t) {
    if (t = t || "float32", t !== "float32" && t !== "int32")
      throw new Fe(`randomNormal does not support dType ${t}.`);
    return Dp(e, this.mean, this.stddev, t, this.seed);
  }
  getConfig() {
    return { mean: this.mean, stddev: this.stddev, seed: this.seed };
  }
};
Ob.className = "RandomNormal";
re.registerClass(Ob);
var Pb = class extends ns {
  constructor(e) {
    super(), this.DEFAULT_MEAN = 0, this.DEFAULT_STDDEV = 0.05, this.mean = e.mean || this.DEFAULT_MEAN, this.stddev = e.stddev || this.DEFAULT_STDDEV, this.seed = e.seed;
  }
  apply(e, t) {
    if (t = t || "float32", t !== "float32" && t !== "int32")
      throw new Fe(`truncatedNormal does not support dType ${t}.`);
    return vb(e, this.mean, this.stddev, t, this.seed);
  }
  getConfig() {
    return { mean: this.mean, stddev: this.stddev, seed: this.seed };
  }
};
Pb.className = "TruncatedNormal";
re.registerClass(Pb);
var zb = class extends ns {
  constructor(e) {
    super(), this.gain = e.gain != null ? e.gain : 1;
  }
  apply(e, t) {
    return j(() => {
      if (e.length !== 2 || e[0] !== e[1])
        throw new G("Identity matrix initializer can only be used for 2D square matrices.");
      return V(this.gain, yS(e[0]));
    });
  }
  getConfig() {
    return { gain: this.gain };
  }
};
zb.className = "Identity";
re.registerClass(zb);
function yz(e, t = "channelsLast") {
  let n, s;
  if (Ct(t), e.length === 2)
    n = e[0], s = e[1];
  else if ([3, 4, 5].indexOf(e.length) !== -1) {
    if (t === "channelsFirst") {
      let r = cr(e, 2);
      n = e[1] * r, s = e[0] * r;
    } else if (t === "channelsLast") {
      let r = cr(e, 0, e.length - 2);
      n = e[e.length - 2] * r, s = e[e.length - 1] * r;
    }
  } else {
    let r = cr(e);
    n = Math.sqrt(r), s = Math.sqrt(r);
  }
  return [n, s];
}
var xn = class extends ns {
  constructor(e) {
    if (super(), e.scale < 0)
      throw new G(`scale must be a positive float. Got: ${e.scale}`);
    this.scale = e.scale == null ? 1 : e.scale, this.mode = e.mode == null ? "fanIn" : e.mode, gz(this.mode), this.distribution = e.distribution == null ? "normal" : e.distribution, bz(this.distribution), this.seed = e.seed;
  }
  apply(e, t) {
    let n = yz(e), s = n[0], r = n[1], a = this.scale;
    if (this.mode === "fanIn" ? a /= Math.max(1, s) : this.mode === "fanOut" ? a /= Math.max(1, r) : a /= Math.max(1, (s + r) / 2), this.distribution === "normal") {
      let i = Math.sqrt(a);
      if (t = t || "float32", t !== "float32" && t !== "int32")
        throw new Fe(`${this.getClassName()} does not support dType ${t}.`);
      return vb(e, 0, i, t, this.seed);
    } else {
      let i = Math.sqrt(3 * a);
      return Vl(e, -i, i, t);
    }
  }
  getConfig() {
    return { scale: this.scale, mode: this.mode, distribution: this.distribution, seed: this.seed };
  }
};
xn.className = "VarianceScaling";
re.registerClass(xn);
var Op = class extends xn {
  constructor(e) {
    super({ scale: 1, mode: "fanAvg", distribution: "uniform", seed: e == null ? null : e.seed });
  }
  getClassName() {
    return xn.className;
  }
};
Op.className = "GlorotUniform";
re.registerClass(Op);
var Pp = class extends xn {
  constructor(e) {
    super({ scale: 1, mode: "fanAvg", distribution: "normal", seed: e == null ? null : e.seed });
  }
  getClassName() {
    return xn.className;
  }
};
Pp.className = "GlorotNormal";
re.registerClass(Pp);
var zp = class extends xn {
  constructor(e) {
    super({ scale: 2, mode: "fanIn", distribution: "normal", seed: e == null ? null : e.seed });
  }
  getClassName() {
    return xn.className;
  }
};
zp.className = "HeNormal";
re.registerClass(zp);
var Mp = class extends xn {
  constructor(e) {
    super({ scale: 2, mode: "fanIn", distribution: "uniform", seed: e == null ? null : e.seed });
  }
  getClassName() {
    return xn.className;
  }
};
Mp.className = "HeUniform";
re.registerClass(Mp);
var Lp = class extends xn {
  constructor(e) {
    super({ scale: 1, mode: "fanIn", distribution: "normal", seed: e == null ? null : e.seed });
  }
  getClassName() {
    return xn.className;
  }
};
Lp.className = "LeCunNormal";
re.registerClass(Lp);
var Bp = class extends xn {
  constructor(e) {
    super({ scale: 1, mode: "fanIn", distribution: "uniform", seed: e == null ? null : e.seed });
  }
  getClassName() {
    return xn.className;
  }
};
Bp.className = "LeCunNormal";
re.registerClass(Bp);
var Mb = class extends ns {
  constructor(e) {
    if (super(), this.DEFAULT_GAIN = 1, this.gain = e.gain == null ? this.DEFAULT_GAIN : e.gain, this.seed = e.seed, this.seed != null)
      throw new Fe("Random seed is not implemented for Orthogonal Initializer yet.");
  }
  apply(e, t) {
    return j(() => {
      if (e.length < 2)
        throw new Fe("Shape must be at least 2D.");
      e[0] * e[1] > 2e3 && console.warn(`Orthogonal initializer is being called on a matrix with more than 2000 (${e[0] * e[1]}) elements: Slowness may result.`);
      let n = e[0] > e[1] ? [e[1], e[0]] : e, s = Dp(n, 0, 1, "float32"), r = JO.gramSchmidt(s);
      return e[0] > e[1] && (r = Ge(r)), V(this.gain, r);
    });
  }
  getConfig() {
    return { gain: this.gain, seed: this.seed };
  }
};
Mb.className = "Orthogonal";
re.registerClass(Mb);
var $x = { constant: "Constant", glorotNormal: "GlorotNormal", glorotUniform: "GlorotUniform", heNormal: "HeNormal", heUniform: "HeUniform", identity: "Identity", leCunNormal: "LeCunNormal", leCunUniform: "LeCunUniform", ones: "Ones", orthogonal: "Orthogonal", randomNormal: "RandomNormal", randomUniform: "RandomUniform", truncatedNormal: "TruncatedNormal", varianceScaling: "VarianceScaling", zeros: "Zeros" };
function _x(e, t = {}) {
  return Wl(e, re.SerializationMap.getMap().classNameMap, t, "initializer");
}
function yt(e) {
  return _b(e);
}
function ft(e) {
  if (typeof e == "string") {
    let t = e in $x ? $x[e] : e;
    if (t === "GlorotNormal")
      return new Pp();
    if (t === "GlorotUniform")
      return new Op();
    if (t === "HeNormal")
      return new zp();
    if (t === "HeUniform")
      return new Mp();
    if (t === "LeCunNormal")
      return new Lp();
    if (t === "LeCunUniform")
      return new Bp();
    {
      let n = {};
      return n.className = t, n.config = {}, _x(n);
    }
  } else
    return e instanceof ns ? e : _x(e);
}
function Cm(e) {
  return Array.isArray(e) && Array.isArray(e[0]);
}
function $d(e) {
  return e.length === 0 ? [] : Array.isArray(e[0]) ? e : [e];
}
function Oe(e) {
  let t;
  if (Array.isArray(e)) {
    if (e.length !== 1)
      throw new G(`Expected Tensor length to be 1; got ${e.length}`);
    t = e[0];
  } else
    t = e;
  return t;
}
function nt(e) {
  if (Array.isArray(e) && Array.isArray(e[0])) {
    if (e.length === 1)
      return e = e, e[0];
    throw new G(`Expected exactly 1 Shape; got ${e.length}`);
  } else
    return e;
}
function _d(e) {
  let t = 0;
  for (let n of e)
    n.shape.length === 0 ? t += 1 : t += n.shape.reduce((s, r) => s * r);
  return t;
}
var Ax = "Variable";
var vz = class {
  constructor(e, t = "float32", n = Ax, s = true, r = null) {
    this.dtype = t == null ? "float32" : t, this.shape = e.shape, this.id = JS(), n = n == null ? Ax : n, this.originalName = tI(n), this.name = nI(this.originalName), this.trainable_ = s, this.constraint = r, this.val = nF(e, this.trainable_, this.name, this.dtype);
  }
  read() {
    return this.assertNotDisposed(), this.val;
  }
  write(e) {
    return this.assertNotDisposed(), xz(this.val, e), this.val.id !== e.id && (this.val.assign(e), this.constraint != null && this.val.assign(this.constraint.apply(this.val))), this;
  }
  dispose() {
    this.assertNotDisposed(), this.val.dispose();
  }
  assertNotDisposed() {
    if (this.val.isDisposed)
      throw new Error(`LayersVariable ${this.name} is already disposed.`);
  }
  get trainable() {
    return this.trainable_;
  }
  set trainable(e) {
    this.trainable_ = e, this.val.trainable = e;
  }
};
function xz(e, t) {
  if (e.shape.toString() !== t.shape.toString())
    throw new Error("Shape mismatch: " + JSON.stringify(e.shape) + " vs. " + JSON.stringify(t.shape));
}
function Nm(e) {
  return e.map((t) => t.read());
}
function Lb(e) {
  e.forEach((t) => {
    t[0].write(t[1]);
  });
}
var Ft = class {
  constructor(e) {
    this.dtype = e.dtype, this.shape = e.shape, e.shape != null ? this.ndim = e.shape.length : this.ndim = e.ndim, this.maxNDim = e.maxNDim, this.minNDim = e.minNDim, this.axes = e.axes || {};
  }
};
var $s = class {
  constructor(e, t, n, s, r, a, i) {
    this.dtype = e, this.shape = t, this.sourceLayer = n, this.inputs = s, this.callArgs = r, this.outputTensorIndex = i, this.id = JS(), a != null && (this.originalName = tI(a), this.name = nI(this.originalName)), this.rank = t.length;
  }
};
var wz = 0;
var Vp = class {
  constructor(e, t) {
    this.callArgs = t, this.id = wz++, this.outboundLayer = e.outboundLayer, this.inboundLayers = e.inboundLayers, this.nodeIndices = e.nodeIndices, this.tensorIndices = e.tensorIndices, this.inputTensors = e.inputTensors, this.outputTensors = e.outputTensors, this.inputMasks = e.inputMasks, this.outputMasks = e.outputMasks, this.inputShapes = e.inputShapes, this.outputShapes = e.outputShapes;
    for (let n of e.inboundLayers)
      n != null && n.outboundNodes.push(this);
    e.outboundLayer.inboundNodes.push(this);
  }
  getConfig() {
    let e = [];
    for (let t of this.inboundLayers)
      t != null ? e.push(t.name) : e.push(null);
    return { outboundLayer: this.outboundLayer ? this.outboundLayer.name : null, inboundLayers: e, nodeIndices: this.nodeIndices, tensorIndices: this.tensorIndices };
  }
};
var kz = 0;
var He = class extends re.Serializable {
  constructor(e = {}) {
    super(), this._callHook = null, this._addedWeightNames = [], this._stateful = false, this.id = kz++, this.activityRegularizer = null, this.inputSpec = null, this.supportsMasking = false, this._trainableWeights = [], this._nonTrainableWeights = [], this._losses = [], this._updates = [], this._built = false, this.inboundNodes = [], this.outboundNodes = [];
    let t = e.name;
    if (!t) {
      let n = this.getClassName();
      t = Vs(n) + "_" + Ep(n);
    }
    if (this.name = t, this.trainable_ = e.trainable == null ? true : e.trainable, e.inputShape != null || e.batchInputShape != null) {
      let n;
      if (e.batchInputShape != null)
        n = e.batchInputShape;
      else if (e.inputShape != null) {
        let r = null;
        e.batchSize != null && (r = e.batchSize), n = [r].concat(e.inputShape);
      }
      this.batchInputShape = n;
      let s = e.dtype;
      s == null && (s = e.inputDType), s == null && (s = "float32"), this.dtype = s;
    }
    e.weights != null ? this.initialWeights = e.weights : this.initialWeights = null, this._refCount = null, this.fastWeightInitDuringBuild = false;
  }
  static nodeKey(e, t) {
    return e.name + "_ib-" + t.toString();
  }
  getNodeAtIndex(e, t) {
    if (this.inboundNodes.length === 0)
      throw new fs(`The layer has never been called and thus has no defined ${t}.`);
    if (this.inboundNodes.length <= e)
      throw new G(`Asked to get ${t} at node ${e}, but the layer has only ${this.inboundNodes.length} inbound nodes.`);
    return this.inboundNodes[e];
  }
  getInputAt(e) {
    return bn(this.getNodeAtIndex(e, "input").inputTensors);
  }
  getOutputAt(e) {
    return bn(this.getNodeAtIndex(e, "output").outputTensors);
  }
  get input() {
    if (this.inboundNodes.length > 1)
      throw new Bs(`Layer ${this.name} has multiple inbound nodes, hence the notion of "layer input" is ill-defined. Use \`getInputAt(nodeIndex)\` instead.`);
    if (this.inboundNodes.length === 0)
      throw new Bs(`Layer ${this.name} is not connected, no input to return.`);
    return bn(this.getNodeAtIndex(0, "input").inputTensors);
  }
  get output() {
    if (this.inboundNodes.length === 0)
      throw new Bs(`Layer ${this.name} has no inbound nodes.`);
    if (this.inboundNodes.length > 1)
      throw new Bs(`Layer ${this.name} has multiple inbound nodes, hence the notion of "layer output" is ill-defined. Use \`getOutputAt(nodeIndex)\` instead.`);
    return bn(this.getNodeAtIndex(0, "output").outputTensors);
  }
  get losses() {
    return this._losses;
  }
  calculateLosses() {
    return this.losses.map((e) => e());
  }
  get updates() {
    return this._updates;
  }
  get built() {
    return this._built;
  }
  set built(e) {
    this._built = e;
  }
  get trainable() {
    return this.trainable_;
  }
  set trainable(e) {
    this._trainableWeights.forEach((t) => t.trainable = e), this.trainable_ = e;
  }
  get trainableWeights() {
    return this.trainable_ ? this._trainableWeights.filter((e) => e.trainable) : [];
  }
  set trainableWeights(e) {
    this._trainableWeights = e;
  }
  get nonTrainableWeights() {
    return this.trainable ? this._trainableWeights.filter((e) => !e.trainable).concat(this._nonTrainableWeights) : this._trainableWeights.concat(this._nonTrainableWeights);
  }
  set nonTrainableWeights(e) {
    this._nonTrainableWeights = e;
  }
  get weights() {
    return this.trainableWeights.concat(this.nonTrainableWeights);
  }
  get stateful() {
    return this._stateful;
  }
  resetStates() {
    if (!this.stateful)
      throw new Error("Cannot call the resetStates() method of a non-stateful Layer object.");
  }
  assertInputCompatibility(e) {
    if (e = ht(e), this.inputSpec == null || this.inputSpec.length === 0)
      return;
    let t = ht(this.inputSpec);
    if (e.length !== t.length)
      throw new G(`Layer ${this.name} expects ${t.length} inputs, but it received ${e.length} input tensors. Input received: ${e}`);
    for (let n = 0; n < e.length; n++) {
      let s = e[n], r = t[n];
      if (r == null)
        continue;
      let a = s.rank;
      if (r.ndim != null && a !== r.ndim)
        throw new G(`Input ${n} is incompatible with layer ${this.name}: expected ndim=${r.ndim}, found ndim=${a}`);
      if (r.maxNDim != null && a > r.maxNDim)
        throw new G(`Input ${n} is incompatible with layer ${this.name}: expected max_ndim=${r.maxNDim}, found ndim=${a}`);
      if (r.minNDim != null && a < r.minNDim)
        throw new G(`Input ${n} is incompatible with layer ${this.name}: expected min_ndim=${r.minNDim}, found ndim=${a}.`);
      if (r.dtype != null && s.dtype !== r.dtype)
        throw new G(`Input ${n} is incompatible with layer ${this.name} : expected dtype=${r.dtype}, found dtype=${s.dtype}.`);
      if (r.axes) {
        let i = s.shape;
        for (let o in r.axes) {
          let u = Number(o), l = r.axes[o], c = u >= 0 ? i[u] : i[i.length + u];
          if (l != null && [l, null].indexOf(c) === -1)
            throw new G(`Input ${n} is incompatible with layer ${this.name}: expected axis ${u} of input shape to have value ${l} but got shape ${i}.`);
        }
      }
      if (r.shape != null)
        for (let i = 0; i < r.shape.length; ++i) {
          let o = r.shape[i], u = s.shape[i];
          if (o != null && u != null && o !== u)
            throw new G(`Input ${n} is incompatible with layer ${this.name}: expected shape=${r.shape}, found shape=${s.shape}.`);
        }
    }
  }
  call(e, t) {
    return e;
  }
  invokeCallHook(e, t) {
    this._callHook != null && this._callHook(e, t);
  }
  setCallHook(e) {
    this._callHook = e;
  }
  clearCallHook() {
    this._callHook = null;
  }
  apply(e, t) {
    t = t || {}, this.assertNotDisposed();
    let n = ht(e), s = true;
    for (let a of n)
      if (!(a instanceof $s)) {
        s = false;
        break;
      }
    let r = true;
    for (let a of n)
      if (a instanceof $s) {
        r = false;
        break;
      }
    if (s === r)
      throw new G("Arguments to apply() must be all SymbolicTensors or all Tensors");
    return ta(this.name, () => {
      if (!this.built) {
        this.assertInputCompatibility(e);
        let a = [];
        for (let i of ht(e))
          a.push(i.shape);
        this.build(bn(a)), this.built = true, this.initialWeights && this.setWeights(this.initialWeights), this._refCount === null && r && (this._refCount = 1);
      }
      if (this.assertInputCompatibility(e), r) {
        let a = this.call(e, t), i = ht(a), o = [];
        for (let u of i)
          n.indexOf(u) !== -1 && (u = u.clone()), o.push(u);
        if (a = bn(o), this.activityRegularizer != null)
          throw new Fe("Layer invocation in the presence of activity regularizer(s) is not supported yet.");
        return a;
      } else {
        let a = Sz(e), i = this.computeOutputShape(a), o, u = Iz(e);
        if (this.warnOnIncompatibleInputShape(Array.isArray(e) ? a[0] : a), i != null && i.length > 0 && Array.isArray(i[0]) ? o = i.map((l, c) => new $s(u, l, this, ht(e), t, this.name, c)) : o = new $s(u, i, this, ht(e), t, this.name), this.addInboundNode(e, o, null, null, a, i, t), this._refCount++, this.activityRegularizer != null)
          throw new Fe("Layer invocation in the presence of activity regularizer(s) is not supported yet.");
        return o;
      }
    });
  }
  warnOnIncompatibleInputShape(e) {
    if (this.batchInputShape != null)
      if (e.length !== this.batchInputShape.length)
        console.warn(`The rank of the input tensor provided (shape: ${JSON.stringify(e)}) does not match that of the batchInputShape (${JSON.stringify(this.batchInputShape)}) of the layer ${this.name}`);
      else {
        let t = false;
        this.batchInputShape.forEach((n, s) => {
          n != null && e[s] != null && e[s] !== n && (t = true);
        }), t && console.warn(`The shape of the input tensor (${JSON.stringify(e)}) does not match the expectation of layer ${this.name}: ${JSON.stringify(this.batchInputShape)}`);
      }
  }
  get outputShape() {
    if (this.inboundNodes == null || this.inboundNodes.length === 0)
      throw new Bs(`The layer ${this.name} has never been called and thus has no defined output shape.`);
    let e = [];
    for (let t of this.inboundNodes) {
      let n = JSON.stringify(t.outputShapes);
      e.indexOf(n) === -1 && e.push(n);
    }
    if (e.length === 1) {
      let t = this.inboundNodes[0].outputShapes;
      return Array.isArray(t) && Array.isArray(t[0]) && t.length === 1 ? t[0] : t;
    } else
      throw new Bs(`The layer ${this.name} has multiple inbound nodes with different output shapes. Hence the notion of "output shape" is ill-defined for the layer.`);
  }
  countParams() {
    if (!this.built)
      throw new fs(`You tried to call countParams() on ${this.name}, but the layer is not built yet. Build it first by calling build(batchInputShape).`);
    return _d(this.weights);
  }
  build(e) {
    this.built = true;
  }
  getWeights(e = false) {
    return Nm(e ? this.trainableWeights : this.weights);
  }
  setWeights(e) {
    j(() => {
      let t = this.weights;
      if (t.length !== e.length)
        throw new G(`You called setWeights(weights) on layer "${this.name}" with a weight list of length ${e.length}, but the layer was expecting ${t.length} weights. Provided weights: ${e}...`);
      if (t.length === 0)
        return;
      let n = [], s = Nm(t);
      for (let r = 0; r < s.length; ++r) {
        let a = s[r], i = t[r], o = e[r];
        if (!w.arraysEqual(a.shape, o.shape))
          throw new G(`Layer weight shape ${a.shape} not compatible with provided weight shape ${o.shape}`);
        n.push([i, o]);
      }
      Lb(n);
    });
  }
  addWeight(e, t, n, s, r, a, i, o) {
    if (this._addedWeightNames.indexOf(e) !== -1)
      throw new G(`Duplicate weight name ${e} for layer ${this.name}`);
    this._addedWeightNames.push(e), n == null && (n = "float32"), this.fastWeightInitDuringBuild && (s = o != null ? o() : ft("zeros"));
    let u = s.apply(t, n), l = new vz(u, n, e, a, i);
    return u.dispose(), r != null && this.addLoss(() => r.apply(l.read())), a == null && (a = true), a ? this._trainableWeights.push(l) : this._nonTrainableWeights.push(l), l;
  }
  setFastWeightInitDuringBuild(e) {
    this.fastWeightInitDuringBuild = e;
  }
  addLoss(e) {
    e == null || Array.isArray(e) && e.length === 0 || (e = ht(e), this._losses !== void 0 && this._losses !== null && this.losses.push(...e));
  }
  computeOutputShape(e) {
    return e;
  }
  computeMask(e, t) {
    if (!this.supportsMasking) {
      if (t != null)
        if (Array.isArray(t))
          t.forEach((n) => {
            if (n != null)
              throw new TypeError(`Layer ${this.name} does not support masking, but was passed an inputMask.`);
          });
        else
          throw new TypeError(`Layer ${this.name} does not support masking, but was passed an inputMask.`);
      return null;
    }
    return t;
  }
  addInboundNode(e, t, n, s, r, a, i = null) {
    let o = ht(e);
    t = ht(t), n = ht(n), s = ht(s), r = $d(r), a = $d(a);
    let u = [], l = [], c = [];
    for (let p of o)
      u.push(p.sourceLayer), l.push(p.nodeIndex), c.push(p.tensorIndex);
    new Vp({ outboundLayer: this, inboundLayers: u, nodeIndices: l, tensorIndices: c, inputTensors: o, outputTensors: t, inputMasks: n, outputMasks: s, inputShapes: r, outputShapes: a }, i);
    for (let p = 0; p < t.length; p++)
      t[p].sourceLayer = this, t[p].nodeIndex = this.inboundNodes.length - 1, t[p].tensorIndex = p;
  }
  getConfig() {
    let e = { name: this.name, trainable: this.trainable };
    return this.batchInputShape != null && (e.batchInputShape = this.batchInputShape), this.dtype != null && (e.dtype = this.dtype), e;
  }
  disposeWeights() {
    return this.weights.forEach((e) => e.dispose()), this.weights.length;
  }
  assertNotDisposed() {
    if (this._refCount === 0)
      throw new Error(`Layer '${this.name}' is already disposed.`);
  }
  dispose() {
    if (!this.built)
      throw new Error(`Cannot dispose Layer ${this.name} because it has not been built yet.`);
    if (this._refCount === null)
      throw new Error(`Cannot dispose Layer ${this.name} because it has not been used yet.`);
    this.assertNotDisposed();
    let e = 0;
    return --this._refCount === 0 && (e = this.disposeWeights()), { refCountAfterDispose: this._refCount, numDisposedVariables: e };
  }
};
function Sz(e) {
  e = ht(e);
  let t = [];
  for (let n of e)
    t.push(n.shape);
  return bn(t);
}
function Iz(e) {
  return "float32";
}
function iI(e, t, n) {
  if ((t == null || n != null && n > 0) && (t = e.sourceLayer, n = e.nodeIndex), t.inboundNodes.length === 0)
    return [e];
  {
    let s = t.inboundNodes[n];
    if (s.inboundLayers.length === 0)
      return s.inputTensors;
    {
      let r = [];
      for (let a = 0; a < s.inboundLayers.length; a++) {
        let i = s.inputTensors[a], o = s.inboundLayers[a], u = s.nodeIndices[a], l = iI(i, o, u);
        for (let c of l)
          r.indexOf(c) === -1 && r.push(c);
      }
      return r;
    }
  }
}
var tu = class extends He {
  constructor(e) {
    if (super({ dtype: e.dtype, name: e.name != null ? e.name : Ep("input").toString() }), e.batchSize == null && (e.batchSize = null), e.sparse == null && (e.sparse = false), this.trainable = false, this.built = true, this.sparse = e.sparse, e.inputShape != null && e.batchInputShape != null)
      throw new G("Only provide the inputShape OR batchInputShape argument to inputLayer, not both at the same time.");
    let t = e.batchInputShape;
    if (t == null) {
      if (e.inputShape == null)
        throw new G("An InputLayer should be passed either a `batchInputShape` or an `inputShape`.");
      t = [e.batchSize].concat(e.inputShape);
    } else if (e.batchSize != null)
      throw new G("Cannot specify batchSize if batchInputShape is specified when creating an InputLayer.");
    let n = e.dtype || "float32";
    this.batchInputShape = t, this.dtype = n, this.inputSpec = [{ shape: t }];
    let s = new $s(this.dtype, this.batchInputShape, this, [], {}, this.name);
    s.nodeIndex = 0, s.tensorIndex = 0, new Vp({ outboundLayer: this, inboundLayers: [], nodeIndices: [], tensorIndices: [], inputTensors: [s], outputTensors: [s], inputMasks: [null], outputMasks: [null], inputShapes: [t], outputShapes: [t] });
  }
  apply(e, t) {
    throw new G(`Cannot pass any input to an InputLayer's apply() method. InputLayer name: ${this.name}`);
  }
  dispose() {
    return { refCountAfterDispose: this._refCount, numDisposedVariables: 0 };
  }
  getConfig() {
    return { batchInputShape: this.batchInputShape, dtype: this.dtype, sparse: this.sparse, name: this.name };
  }
};
tu.className = "InputLayer";
re.registerClass(tu);
function oI(e) {
  if (e.batchShape == null && e.shape == null)
    throw new Error("Please provide to Input either a `shape` or a `batchShape` argument. Note that `shape` does not include the batch dimension.");
  if (e.batchShape != null && e.shape != null)
    throw new G("Please provide either a `shape` or `batchShape` argument to Input, but not both.");
  let t = e.batchShape;
  e.shape != null && t == null && (t = [null].concat(e.shape));
  let n = e.dtype;
  return n == null && (n = "float32"), new tu({ batchInputShape: t, name: e.name, dtype: n, sparse: e.sparse }).inboundNodes[0].outputTensors[0];
}
function Cz(e, t) {
  if (e.dtype == null || e.dtype === t.dtype)
    return t;
  try {
    return le(t, e.dtype);
  } catch (n) {
    throw new G(`The dtype of the feed (${t.dtype}) can not be cast to the dtype of the key '${e.name}' (${e.dtype}).`);
  }
}
var Zr = class {
  constructor(e) {
    if (this.id2Value = {}, this.id2Mask = {}, this.name2Id = {}, e instanceof Zr)
      for (let t in e.id2Value)
        this.id2Value[t] = e.id2Value[t], t in e.id2Mask && (this.id2Mask[t] = e.id2Mask[t]);
    else {
      if (e == null)
        return;
      for (let t of e)
        this.add(t.key, t.value);
    }
  }
  add(e, t, n) {
    if (this.id2Value[e.id] == null)
      this.id2Value[e.id] = Cz(e, t), this.name2Id[e.name] = e.id, n != null && (this.id2Mask[e.id] = n);
    else
      throw new G(`Duplicate key: name=${e.name}, id=${e.id}`);
    return this;
  }
  addFeed(e) {
    this.add(e.key, e.value);
  }
  hasKey(e) {
    return this.id2Value[e.id] != null;
  }
  names() {
    return Object.keys(this.name2Id);
  }
  getValue(e) {
    if (e instanceof $s) {
      if (this.id2Value[e.id] == null)
        throw new G(`Nonexistent key: ${e.name}`);
      return this.id2Value[e.id];
    } else {
      let t = this.name2Id[e];
      if (t == null)
        throw new G(`Feed dict has no SymbolicTensor name: ${e}`);
      return this.id2Value[t];
    }
  }
  getMask(e) {
    if (e instanceof $s) {
      if (this.id2Value[e.id] == null)
        throw new G(`Nonexistent key: ${e.name}`);
      return this.id2Mask[e.id];
    } else {
      let t = this.name2Id[e];
      if (t == null)
        throw new G(`Feed dict has no SymbolicTensor name: ${e}`);
      return this.id2Mask[t];
    }
  }
  disposeMasks() {
    this.id2Mask != null && De(this.id2Mask);
  }
};
var Ad = new YS();
var Ed = new YS();
function Nz(e) {
  Ad != null && Ad.setMaxEntries(e), Ed != null && Ed.setMaxEntries(e);
}
function Fu(e, t, n, s) {
  let r = n == null ? false : n.training, a = Array.isArray(e), i = a ? e : [e], o = i.map((f) => f.name), u = [], l = t.names();
  for (let f of o)
    l.indexOf(f) !== -1 ? u.push(t.getValue(f)) : u.push(null);
  s != null && (s.maxNumTensors = -1 / 0, s.minNumTensors = 1 / 0);
  let c = o.join(",") + "|" + t.names().sort().join(","), p = Ad.get(c), d;
  if (p == null) {
    let f = Tz(i, t);
    p = f.sorted, d = f.recipientCounts, Ad.put(c, p), Ed.put(c, d);
  }
  d = {}, r || Object.assign(d, Ed.get(c));
  let h = new Zr(t);
  for (let f = 0; f < p.length; ++f) {
    if (s != null) {
      let E = bm().numTensors;
      E > s.maxNumTensors && (s.maxNumTensors = E), E < s.minNumTensors && (s.minNumTensors = E);
    }
    let m = p[f], g = m.sourceLayer;
    if (g instanceof tu)
      continue;
    let b = [], y = [], v = [], x = false;
    for (let E of m.inputs) {
      let P = h.getValue(E), A = h.getMask(E);
      b.push(P), y.push(A), A != null && (x = true), r || (d[E.name]--, d[E.name] === 0 && !t.hasKey(E) && o.indexOf(E.name) === -1 && !P.isDisposed && E.sourceLayer.stateful !== true && v.push(P));
    }
    x && (n = n || {}, n.mask = y[0]);
    let k = ht(g.apply(b, n)), I = null;
    g.supportsMasking && (I = g.computeMask(b, y));
    let $ = _z(m), R = Array.isArray($) ? $ : [$];
    for (let E = 0; E < R.length; ++E) {
      h.hasKey(R[E]) || h.add(R[E], k[E], Array.isArray(I) ? I[0] : I);
      let P = o.indexOf(R[E].name);
      P !== -1 && (u[P] = k[E]);
    }
    r || De(v);
  }
  return h.disposeMasks(), a ? u : u[0];
}
function Tz(e, t) {
  w.assert(e != null && e.length > 0, () => "Expected at least one fetch, got none");
  let n = [], s = {};
  if (e.length === 1) {
    let r = Ex(e[0], t);
    n = r.sorted, s = r.recipientMap;
  } else {
    let r = /* @__PURE__ */ new Set();
    for (let a of e) {
      let { sorted: i, recipientMap: o } = Ex(a, t);
      for (let u of i)
        r.has(u.name) || (n.push(u), r.add(u.name));
      for (let u in o)
        s[u] == null && (s[u] = /* @__PURE__ */ new Set()), o[u].forEach((l) => s[u].add(l));
    }
  }
  return { sorted: n, recipientCounts: $z(s) };
}
function $z(e) {
  let t = {};
  for (let n in e)
    t[n] = e[n].size;
  return t;
}
function Ex(e, t) {
  let n = /* @__PURE__ */ new Set(), s = [], r = {};
  for (let o of t.names())
    n.add(o);
  let a = [], i = [];
  for (a.push(e); a.length > 0; ) {
    let o = a[a.length - 1];
    if (n.has(o.name)) {
      a.pop();
      continue;
    }
    let u = i[i.length - 1] === a.length - 1;
    if (o.inputs.length === 0 || u)
      a.pop(), s.push(o), n.add(o.name), u && i.pop();
    else {
      i.push(a.length - 1);
      for (let l of o.inputs)
        r[l.name] == null && (r[l.name] = /* @__PURE__ */ new Set()), r[l.name].add(o.name), !n.has(l.name) && a.push(l);
    }
  }
  return { sorted: s, recipientMap: r };
}
function _z(e) {
  let t;
  if (e.sourceLayer.inboundNodes.length === 1)
    t = e.sourceLayer.output;
  else {
    let n = null;
    for (let s = 0; s < e.sourceLayer.inboundNodes.length; ++s)
      for (let r of e.sourceLayer.inboundNodes[s].outputTensors)
        if (r.id === e.id) {
          n = s;
          break;
        }
    t = e.sourceLayer.getOutputAt(n);
  }
  return t;
}
var Az = K();
Az.registerFlag("TOPOLOGICAL_SORT_CACHE_MAX_ENTRIES", () => 100, Nz);
var uI = { kernelName: po, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(e, Cp(le(n, "float32"), -1)) };
} };
var Ez = { kernelName: ol, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => {
    let s = ct(le(n, "float32")), r = dn(ge(we(1), s));
    return kt(xe(e, r));
  } };
} };
var Rz = { kernelName: ul, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => {
    let s = dn(ge(ct(le(n, "float32")), 1));
    return xe(e, s);
  } };
} };
var Dz = { kernelName: Sr, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t, r = rt(n.shape, s.shape);
  return { a: () => {
    let o = e, u = At(n.shape, r);
    return u.length > 0 && (o = ve(o, u)), U(o, n.shape);
  }, b: () => {
    let o = e, u = At(s.shape, r);
    return u.length > 0 && (o = ve(o, u)), U(o, s.shape);
  } };
} };
var Fz = { kernelName: Sa, saveAllInputs: true, gradFunc: (e, t) => {
  let n = {};
  return t.forEach((s, r) => {
    n[r] = () => e.clone();
  }), n;
} };
var Oz = { kernelName: Ia, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => je(n) };
} };
var Pz = { kernelName: dl, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => je(n) };
} };
var zz = { kernelName: pl, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => xe(e, dn(ge(we(1), ct(le(n, "float32"))))) };
} };
var Mz = { kernelName: hl, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => {
    let s = dn(ie(we(1), ct(le(n, "float32"))));
    return xe(e, s);
  } };
} };
var Lz = { kernelName: gl, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t, r = rt(n.shape, s.shape);
  return { a: () => {
    let o = ie(ct(n), ct(s)), u = V(e, xe(s, o)), l = At(n.shape, r);
    return l.length > 0 && (u = ve(u, l)), U(u, n.shape);
  }, b: () => {
    let o = ie(ct(n), ct(s)), u = kt(V(e, xe(n, o))), l = At(s.shape, r);
    return l.length > 0 && (u = ve(u, l)), U(u, s.shape);
  } };
} };
var Bz = { kernelName: fl, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => xe(e, ie(ct(le(n, "float32")), 1)) };
} };
var Vz = { kernelName: ml, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => xe(e, ge(we(1), ct(le(n, "float32")))) };
} };
function Wz(e, t, n, s, r, a) {
  let i = _(e, "dy", "avgPool3dGrad"), o = _(t, "input", "avgPool3dGrad"), u = i, l = o, c = false;
  o.rank === 4 && (c = true, u = U(i, [1, i.shape[0], i.shape[1], i.shape[2], i.shape[3]]), l = U(o, [1, o.shape[0], o.shape[1], o.shape[2], o.shape[3]])), F(u.rank === 5, () => `Error in avgPool3dGrad: dy must be rank 5 but got rank ${u.rank}.`), F(l.rank === 5, () => `Error in avgPool3dGrad: input must be rank 5 but got rank ${l.rank}.`), hn("avgPool3dGrad", r, a);
  let p = { dy: u, input: l }, d = { filterSize: n, strides: s, pad: r, dimRoundingMode: a }, h = M.runKernel(hg, p, d);
  return c ? U(h, [h.shape[1], h.shape[2], h.shape[3], h.shape[4]]) : h;
}
var Uz = L({ avgPool3dGrad_: Wz });
var Gz = { kernelName: Jd, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { filterSize: r, strides: a, pad: i, dimRoundingMode: o } = n;
  return { x: () => Uz(e, s, r, a, i, o) };
} };
function Hz(e, t, n, s, r) {
  let a = _(e, "dy", "avgPoolGrad"), i = _(t, "input", "avgPoolGrad");
  F(i.rank === a.rank, () => `Rank of input (${i.rank}) does not match rank of dy (${a.rank})`);
  let o = i, u = a, l = false;
  i.rank === 3 && (l = true, o = U(i, [1, i.shape[0], i.shape[1], i.shape[2]]), u = U(a, [1, a.shape[0], a.shape[1], a.shape[2]])), F(u.rank === 4, () => `Error in avgPoolGrad: dy must be rank 4 but got rank ${u.rank}.`), F(o.rank === 4, () => `Error in avgPoolGrad: input must be rank 4 but got rank ${o.rank}.`);
  let c = { dy: u, input: o }, p = { filterSize: n, strides: s, pad: r }, d = M.runKernel(pg, c, p);
  return l ? U(d, [d.shape[1], d.shape[2], d.shape[3]]) : d;
}
var qz = L({ avgPoolGrad_: Hz });
var jz = { kernelName: Ca, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { filterSize: r, strides: a, pad: i } = n;
  return { x: () => qz(e, s, r, a, i) };
} };
var Kz = { kernelName: Na, inputsToSave: ["a", "b"], gradFunc: (e, t, n) => {
  let [s, r] = t, { transposeA: a, transposeB: i } = n;
  return !a && !i ? { a: () => Ve(e, r, false, true), b: () => Ve(s, e, true, false) } : !a && i ? { a: () => Ve(e, r, false, false), b: () => Ve(e, s, true, false) } : a && !i ? { a: () => Ve(r, e, false, true), b: () => Ve(s, e, false, false) } : { a: () => Ve(r, e, true, true), b: () => Ve(e, s, true, true) };
} };
var Xz = { kernelName: ho, gradFunc: (e, t, n) => {
  let { blockShape: s, crops: r } = n;
  return { x: () => cb(e, s, r) };
} };
var Yz = { kernelName: M$, gradFunc: (e, t, n) => {
  let s = n, r = s.inputShape, a = s.shape, i = Array.from(a);
  for (let u = r.length - 1; u >= 0; u--)
    if (r[u] === a[u])
      i[u] = 1;
    else if (r[u] !== 1)
      throw new Error(`broadcastTo(): [${r}] cannot be broadcast to [${a}].`);
  let o = [];
  for (let u = 0; u < i.length; u++)
    i[u] > 1 && o.push(u);
  return { x: () => ve(e, o, true) };
} };
var Qz = { kernelName: Ta, gradFunc: (e) => ({ x: () => e.clone() }) };
var Zz = { kernelName: $a, gradFunc: (e) => ({ x: () => je(e) }) };
var Jz = { kernelName: Ir, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { clipValueMin: r, clipValueMax: a } = n;
  return { x: () => vn(Ds(Zo(s, r), Jo(s, a)), e, je(e)) };
} };
var eM = { kernelName: tp, inputsToSave: ["x"], gradFunc: uI.gradFunc };
var tM = { kernelName: fo, saveAllInputs: true, gradFunc: (e, t, n) => {
  let s = t.map((u) => u.shape), { axis: r } = n, a = ts(r, t[0].shape)[0], i = s.map((u) => u[a]);
  return Bn(e, i, a).map((u) => () => u);
} };
var nM = { kernelName: _a, inputsToSave: ["x", "filter"], gradFunc: (e, t, n) => {
  let [s, r] = t, { dilations: a, strides: i, pad: o, dataFormat: u } = n;
  return F(fr(a), () => `Error in gradient of conv2D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${a}'`), { x: () => Jg(s.shape, e, r, i, o, u), filter: () => xb(s, e, r.shape, i, o, u) };
} };
var sM = { kernelName: Aa, inputsToSave: ["dy", "filter"], gradFunc: (e, t, n) => {
  let [s, r] = t, { strides: a, pad: i, dataFormat: o, dimRoundingMode: u } = n;
  return { dy: () => da(e, r, a, i, o, 1, u), filter: () => xb(e, s, r.shape, a, i, o, u) };
} };
function rM(e, t, n, s, r) {
  let a = e;
  e.rank === 4 && (a = U(e, [1, e.shape[0], e.shape[1], e.shape[2], e.shape[3]]));
  let i = t;
  i.rank === 4 && (i = U(t, [1, t.shape[0], t.shape[1], t.shape[2], t.shape[3]])), F(a.rank === 5, () => `Error in conv3dDerFilter: input must be rank 5, but got shape ${a.shape}.`), F(i.rank === 5, () => `Error in conv3dDerFilter: dy must be rank 5, but got shape ${i.shape}.`), F(n.length === 5, () => `Error in conv3dDerFilter: filterShape must be length 5, but got ${n}.`), F(a.shape[4] === n[3], () => `Error in conv3dDerFilter: depth of input ${a.shape[4]}) must match input depth in filter (${n[3]}.`), F(i.shape[4] === n[4], () => `Error in conv3dDerFilter: depth of dy (${i.shape[4]}) must match output depth for filter (${n[4]}).`);
  let o = { x: a, dy: i }, u = { strides: s, pad: r, filterShape: n };
  return M.runKernel(bg, o, u);
}
var aM = L({ conv3DBackpropFilter_: rM });
var iM = { kernelName: np, inputsToSave: ["x", "filter"], gradFunc: (e, t, n) => {
  let { dilations: s, strides: r, pad: a } = n;
  F(fr(s), () => `Error in gradient of conv3D: dilation rates greater than 1 are not yet supported in gradients. Got dilations '${s}'`);
  let [i, o] = t;
  return { x: () => dS(i.shape, e, o, r, a), filter: () => aM(i, e, o.shape, r, a) };
} };
var oM = { kernelName: Ea, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(kt(_S(le(n, "float32"))), e) };
} };
var uM = { kernelName: Ra, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(AS(le(n, "float32")), e) };
} };
var lM = { kernelName: Da, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { axis: r, exclusive: a, reverse: i } = n;
  return { x: () => {
    let o = gS([r], s.rank), u = hS(e, r, a, !i);
    return o != null && (u = Ge(u, o)), u;
  } };
} };
var cM = { kernelName: Fa, inputsToSave: ["x", "filter"], gradFunc: (e, t, n) => {
  let { dilations: s, strides: r, pad: a, dimRoundingMode: i } = n, o = s == null ? [1, 1] : s;
  F(fr(o), () => `Error in gradient of depthwiseConv2dNative: dilation rates greater than 1 are not yet supported. Got dilations '${o}'`);
  let [u, l] = t;
  return F(u.rank === 4, () => `Error in gradient of depthwiseConv2dNative: input must be rank 4, but got rank ${u.rank}.`), F(l.rank === 4, () => `Error in gradient of depthwiseConv2dNative: filter must be rank 4, but got rank ${l.rank}.`), F(u.shape[3] === l.shape[2], () => `Error in gradient of depthwiseConv2d: number of input channels (${u.shape[3]}) must match the inChannels dimension in filter ${l.shape[2]}.`), F(Ps(r, o), () => `Error in gradient of depthwiseConv2d: Either strides or dilations must be  1. Got strides ${r} and dilations '${o}'.`), hn("depthwiseConv2d", a, i), { x: () => LS(u.shape, e, l, r, a, o, i), filter: () => MS(u, e, l.shape, r, a, o, i) };
} };
var dM = { kernelName: sp, inputsToSave: ["x", "filter"], gradFunc: (e, t, n) => {
  let [s, r] = t, a = { x: s, filter: r, dy: e }, i = { x: s, filter: r, dy: e };
  return { x: () => M.runKernel(nm, a, n), filter: () => M.runKernel(sm, i, n) };
} };
var pM = { kernelName: Pa, outputsToSave: [true], gradFunc: (e, t) => {
  let [n] = t, s = { dy: e, y: n };
  return { x: () => M.runKernel(Sg, s) };
} };
var hM = { kernelName: bl, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t, s = V(Yn(kt(ct(n))), 2 / Math.sqrt(Math.PI));
  return { x: () => V(e, s) };
} };
var fM = { kernelName: za, outputsToSave: [true], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(e, n) };
} };
var mM = { kernelName: vo, inputsToSave: ["input"], gradFunc: (e, t) => {
  let [n] = t;
  return { input: () => U(e, n.shape) };
} };
var gM = { kernelName: xo, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(e, Yn(n)) };
} };
var bM = { kernelName: Ma, gradFunc: (e) => ({ x: () => je(e) }) };
var yM = { kernelName: La, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t, r = rt(n.shape, s.shape);
  return { a: () => {
    let o = xe(e, le(s, "float32")), u = At(n.shape, r);
    return u.length > 0 ? U(ve(o, u), n.shape) : o;
  }, b: () => {
    let o = V(e, le(n, "float32")), u = At(s.shape, r);
    u.length > 0 && (o = U(ve(o, u), s.shape));
    let l = ct(s);
    return kt(xe(o, le(l, "float32")));
  } };
} };
var vM = { kernelName: Ba, inputsToSave: ["x", "mean", "variance", "scale"], gradFunc: (e, t, n) => {
  let { varianceEpsilon: s } = n, [r, a, i, o] = t, u = o == null ? we(1) : o, l = At(a.shape, r.shape), c = [];
  if (a.rank === 1) {
    for (let x = 0; x < r.shape.length - 1; ++x)
      c.push(r.shape[x]);
    c.push(1);
  }
  let p = ge(r, a), d = V(e, u), h = TS(ie(i, we(s))), f = V(V(V(h, h), h), we(-0.5));
  return { x: () => a.rank === 1 ? U(V(V(e, hs(U(h, [1, 1, 1, a.shape[0]]), c)), u), r.shape) : U(V(V(e, h), u), r.shape), mean: () => {
    let x = V(V(h, we(-1)), d);
    return a.rank === 1 && (x = ve(x, l)), U(x, a.shape);
  }, variance: () => {
    let x = V(V(f, p), d);
    return a.rank === 1 && (x = ve(x, l)), U(x, a.shape);
  }, scale: () => {
    let x = V(p, h), k = V(e, x);
    return a.rank === 1 && (k = ve(k, l)), U(k, a.shape);
  }, offset: () => {
    let x = e;
    return a.rank === 1 && (x = ve(x, l)), U(x, a.shape);
  } };
} };
var xM = { kernelName: ko, inputsToSave: ["x", "indices"], gradFunc: (e, t, n) => {
  let [s, r] = t, { axis: a } = n, i = ts(a, s.shape)[0];
  return { x: () => {
    let u = s.shape, l = r.size, c = u.slice(0, i), p = c.length, d = u.slice(a, u.length).slice(1), h = d.length, f = Rx(0, p), m = Rx(p + 1, p + 1 + h), g = Dx([c, [l], d]), b = U(e, g), y = U(r, [l]), v = Dx([[p], f, m]), x = Ge(b, v), k = J3(x, y, s.shape[i]), I = nb(v);
    return k = Ge(k, I), k;
  }, indices: () => r };
} };
function Rx(e, t) {
  let n = [];
  for (let s = e; s < t; ++s)
    n.push(s);
  return n;
}
function Dx(e) {
  let t = [];
  for (let n = 0; n < e.length; ++n)
    for (let s = 0; s < e[n].length; ++s)
      t.push(e[n][s]);
  return t;
}
var wM = { kernelName: Va, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t;
  return { a: () => je(n), b: () => je(s) };
} };
var kM = { kernelName: Wa, gradFunc: (e) => ({ x: () => le(e, "float32") }) };
var SM = { kernelName: vl, gradFunc: (e) => ({ x: () => je(e) }) };
var IM = { kernelName: xl, gradFunc: (e) => ({ x: () => je(e) }) };
var CM = { kernelName: wl, gradFunc: (e) => ({ x: () => je(e) }) };
var NM = { kernelName: Ua, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { alpha: r } = n, a = Un(s, 0);
  return { x: () => vn(a, e, V(e, r)) };
} };
var TM = { kernelName: kl, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => xe(e, ie(n, 1)) };
} };
var $M = { kernelName: Ga, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => xe(e, le(n, "float32")) };
} };
var _M = { kernelName: L$, inputsToSave: [], outputsToSave: [true], gradFunc: (e, t, n) => {
  let [s] = t, { axis: r } = n;
  return { logits: () => {
    let i = Yn(s);
    return ge(e, V(ve(e, r, true), i));
  } };
} };
function AM(e, t, n, s = 5, r = 1, a = 1, i = 0.5) {
  let o = { x: e, y: t, dy: n }, u = { depthRadius: s, bias: r, alpha: a, beta: i };
  return M.runKernel(Tg, o, u);
}
var EM = L({ localResponseNormalizationBackprop_: AM });
var RM = { kernelName: op, inputsToSave: ["x"], outputsToSave: [true], gradFunc: (e, t, n) => {
  let [s, r] = t, { depthRadius: a, bias: i, alpha: o, beta: u } = n;
  return { x: () => EM(s, r, e, a, i, o, u) };
} };
function lI(e, t, n, s) {
  return t.rank < n.rank && (t = U(t, pa(t.shape, s))), e.rank < n.rank && (e = U(e, pa(e.shape, s))), { x: () => V(e, le(Xn(n, t), e.dtype)) };
}
var Fx = { kernelName: Ha, inputsToSave: ["x"], outputsToSave: [true], gradFunc: (e, t, n) => {
  let s = n, { reductionIndices: r } = s, a = t[0], i = t[1], o = ts(r, a.shape), u = lI(e, i, a, o);
  return { x: () => u.x() };
} };
var DM = { kernelName: qa, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t;
  return { a: () => V(e, le(Zo(n, s), "float32")), b: () => V(e, le(vS(n, s), "float32")) };
} };
function FM(e, t, n, s, r, a, i) {
  let o = _(e, "dy", "maxPool3dGrad"), u = _(t, "input", "maxPool3dGrad"), l = _(n, "output", "maxPool3dGrad"), c = o, p = u, d = l, h = false;
  u.rank === 4 && (h = true, c = U(o, [1, o.shape[0], o.shape[1], o.shape[2], o.shape[3]]), p = U(u, [1, u.shape[0], u.shape[1], u.shape[2], u.shape[3]]), d = U(l, [1, l.shape[0], l.shape[1], l.shape[2], l.shape[3]])), F(c.rank === 5, () => `Error in maxPool3dGrad: dy must be rank 5 but got rank ${c.rank}.`), F(p.rank === 5, () => `Error in maxPool3dGrad: input must be rank 5 but got rank ${p.rank}.`), F(d.rank === 5, () => `Error in maxPool3dGrad: output must be rank 5 but got rank ${d.rank}.`), hn("maxPool3dGrad", a, i);
  let f = { dy: c, input: p, output: d }, m = { filterSize: s, strides: r, pad: a, dimRoundingMode: i }, g = M.runKernel(_g, f, m);
  return h ? U(g, [g.shape[1], g.shape[2], g.shape[3], g.shape[4]]) : g;
}
var OM = L({ maxPool3dGrad_: FM });
var PM = { kernelName: up, inputsToSave: ["x"], outputsToSave: [true], gradFunc: (e, t, n) => {
  let [s, r] = t, { filterSize: a, strides: i, pad: o, dimRoundingMode: u } = n;
  return { x: () => OM(e, s, r, a, i, o, u) };
} };
function zM(e, t, n, s, r, a, i) {
  let o = _(e, "dy", "maxPoolGrad"), u = _(t, "input", "maxPoolGrad"), l = _(n, "output", "maxPoolGrad");
  F(u.rank === o.rank, () => `Rank of input (${u.rank}) does not match rank of dy (${o.rank})`), F(o.rank === 4, () => `Error in maxPoolGrad: dy must be rank 4 but got rank ${o.rank}.`), F(u.rank === 4, () => `Error in maxPoolGrad: input must be rank 4 but got rank ${u.rank}.`), hn("maxPoolGrad", a, i);
  let c = { dy: o, input: u, output: l }, p = { filterSize: s, strides: r, pad: a, dimRoundingMode: i };
  return M.runKernel($g, c, p);
}
var MM = L({ maxPoolGrad_: zM });
var LM = { kernelName: ja, inputsToSave: ["x"], outputsToSave: [true], gradFunc: (e, t, n) => {
  let [s, r] = t, { filterSize: a, strides: i, pad: o } = n;
  return { x: () => MM(e, s, r, a, i, o) };
} };
var BM = { kernelName: Ka, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { axis: r } = n, a = ts(r, s.shape), o = mS(s.shape, a)[1], u = dt(o);
  return { x: () => {
    let c = s.shape.slice();
    a.forEach((h) => {
      c[h] = 1;
    });
    let p = U(e, c);
    return xe(V(p, Mn(s.shape, "float32")), u);
  } };
} };
var VM = { kernelName: Xa, inputsToSave: ["x"], outputsToSave: [true], gradFunc: (e, t, n) => {
  let s = n, { axis: r } = s, [a, i] = t, o = ts(r, a.shape), u = lI(e, i, a, o);
  return { x: () => u.x() };
} };
var WM = { kernelName: Ya, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t;
  return { a: () => V(e, le(Jo(n, s), "float32")), b: () => V(e, le(Un(n, s), "float32")) };
} };
var UM = { kernelName: Qa, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let s = t[0], { paddings: r } = n, a = r.map((i) => i[0]);
  return { x: () => qe(e, a, s.shape) };
} };
var GM = { kernelName: Il, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t, r = rt(n.shape, s.shape);
  return { a: () => {
    let o = At(n.shape, r);
    return o.length > 0 ? U(ve(e, o), n.shape) : e;
  }, b: () => {
    let o = V(e, kt(kp(xe(n, s)))), u = At(s.shape, r);
    return u.length > 0 ? U(ve(o, u), s.shape) : o;
  } };
} };
var HM = { kernelName: Za, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t, r = rt(n.shape, s.shape);
  return { a: () => {
    let o = V(e, le(s, "float32")), u = At(n.shape, r);
    return u.length > 0 ? U(ve(o, u), n.shape) : o;
  }, b: () => {
    let o = V(e, le(n, "float32")), u = At(s.shape, r);
    return u.length > 0 ? U(ve(o, u), s.shape) : o;
  } };
} };
var qM = { kernelName: $o, gradFunc: (e) => ({ x: () => kt(e) }) };
var jM = { kernelName: Do, inputsToSave: ["indices"], gradFunc: (e, t) => {
  let n = t[0];
  return { indices: () => $t(n.shape, "float32") };
} };
var KM = { kernelName: Ro, gradFunc: (e) => ({ x: () => je(e) }) };
var XM = { kernelName: Fo, saveAllInputs: true, gradFunc: (e, t, n) => {
  let { axis: s } = n;
  return Fs(e, s).map((a) => () => a);
} };
var Ox = { kernelName: Ja, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let s = t[0], { paddings: r } = n, a = r.map((i) => i[0]);
  return { x: () => qe(e, a, s.shape) };
} };
var YM = { kernelName: ei, inputsToSave: ["a", "b"], outputsToSave: [true], gradFunc: (e, t) => {
  let [n, s, r] = t, a = n, i = s, o = rt(a.shape, i.shape);
  return { a: () => {
    let c = le(i, "float32"), p = V(e, V(c, ha(a, ge(c, we(1))))), d = At(a.shape, o);
    return d.length > 0 && (p = ve(p, d)), U(p, a.shape);
  }, b: () => {
    let c = Un(a, 0), p = vn(c, Qn(a), je(a)), d = V(e, V(r, p)), h = At(i.shape, o);
    return h.length > 0 && (d = ve(d, h)), U(d, i.shape);
  } };
} };
var QM = { kernelName: ti, inputsToSave: ["x", "alpha"], gradFunc: (e, t) => {
  let [n, s] = t, r = Un(n, 0);
  return { x: () => vn(r, e, V(e, s)), alpha: () => {
    let a = vn(r, je(e), V(e, n)), i = At(s.shape, e.shape);
    return i.length > 0 && (a = ve(a, i)), U(a, s.shape);
  } };
} };
function ZM(e, t, n) {
  let s = e.shape.slice();
  s[n] = 1;
  let r = U(t, s), a = xm(e, n, true, false), i = xm(e, n, true, true), o = V(a, i);
  return V(r, o);
}
function JM(e, t, n) {
  let s = e.shape.length, r = s - n.length, a = C.getAxesPermutation(n, s), i = e;
  a != null && (i = Ge(e, a));
  let o = i.shape.slice(), l = o.splice(s - n.length, n.length).reduce((d, h) => d * h, 1);
  o.push(l);
  let c = i.reshape(o), p = ZM(c, t, r);
  if (p = p.reshape(i.shape), a != null) {
    let d = C.getUndoAxesPermutation(a);
    p = Ge(p, d);
  }
  return p;
}
var eL = { kernelName: ni, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { axis: r } = n, a = [];
  return r == null ? a = s.shape.map((i, o) => o) : typeof r == "number" ? a = [r] : a = r, { x: () => JM(s, e, a) };
} };
var tL = { kernelName: Oa, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t, r = rt(n.shape, s.shape);
  return { a: () => {
    let o = xe(e, le(s, "float32")), u = At(n.shape, r);
    return u.length > 0 ? U(ve(o, u), n.shape) : o;
  }, b: () => {
    let o = V(e, le(n, "float32")), u = At(s.shape, r);
    u.length > 0 && (o = U(ve(o, u), s.shape));
    let l = ct(s);
    return kt(xe(o, le(l, "float32")));
  } };
} };
var nL = { kernelName: Tl, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => xe(e, kt(ct(n))) };
} };
var sL = { kernelName: ai, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t, s = V(Jo(n, 6), Cp(n));
  return { x: () => V(e, le(s, "float32")) };
} };
var rL = { kernelName: si, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(e, le(Cp(n), "float32")) };
} };
var aL = { kernelName: Oo, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => U(e, n.shape) };
} };
var iL = { kernelName: ri, inputsToSave: ["images"], gradFunc: (e, t, n) => {
  let [s] = t, r = { dy: e, images: s };
  return { images: () => M.runKernel(Dg, r, n) };
} };
var oL = { kernelName: $l, inputsToSave: ["images"], gradFunc: (e, t, n) => {
  let [s] = t, r = { dy: e, images: s };
  return { images: () => M.runKernel(Rg, r, n) };
} };
var uL = { kernelName: Po, gradFunc: (e, t, n) => {
  let { dims: s } = n, r = ts(s, e.shape);
  return { x: () => Jn(e, r) };
} };
var lL = { kernelName: zo, gradFunc: (e) => ({ x: () => je(e) }) };
var cL = { kernelName: ii, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => kt(xe(e, V(ha(n, 1.5), 2))) };
} };
var dL = { kernelName: Lo, inputsToSave: ["condition"], gradFunc: (e, t) => {
  let [n] = t;
  return { condition: () => le(je(n), "float32"), t: () => V(e, le(n, e.dtype)), e: () => V(e, le(ob(n), e.dtype)) };
} };
var pL = { kernelName: _l, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => {
    let s = Un(n, we(0)), r = we(qS), a = we(jS), i = V(e, a), o = V(V(e, r), Yn(le(n, "float32")));
    return vn(s, i, o);
  } };
} };
var hL = { kernelName: ui, outputsToSave: [true], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(e, V(n, ge(we(1), n))) };
} };
var fL = { kernelName: Al, gradFunc: (e) => ({ x: () => je(e) }) };
var mL = { kernelName: oi, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(eb(le(n, "float32")), e) };
} };
var gL = { kernelName: Vo, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(pS(le(n, "float32")), e) };
} };
var bL = { kernelName: Bo, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { begin: r, size: a } = n, i = s.shape, [o, u] = Yk(s, r, a), l = [];
  for (let c = 0; c < e.rank; c++)
    l.push([o[c], i[c] - o[c] - u[c]]);
  return { x: () => bi(e, l) };
} };
var yL = { kernelName: di, outputsToSave: [true], gradFunc: (e, t, n) => {
  let [s] = t, { dim: r } = n, a = true, i = V(e, s);
  return { logits: () => ge(i, V(ve(i, [r], a), s)) };
} };
var vL = { kernelName: El, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(e, Hs(n)) };
} };
var Px = { kernelName: Wo, gradFunc: (e, t, n) => {
  let { blockShape: s, paddings: r } = n;
  return { x: () => Zg(e, s, r) };
} };
var zx = { kernelName: Uo, gradFunc: (e, t, n) => {
  let { axis: s } = n;
  return { x: () => Ot(e, s) };
} };
var xL = { kernelName: li, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => xe(e, V(dn(le(n, "float32")), 2)) };
} };
var wL = { kernelName: Dl, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(e, V(le(n, "float32"), 2)) };
} };
var kL = { kernelName: pi, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t, r = we(2);
  return { a: () => V(e, V(r, ge(n, s))), b: () => V(e, V(r, ge(s, n))) };
} };
var SL = { kernelName: gi, gradFunc: (e) => ({ x: () => je(e) }) };
var IL = { kernelName: hi, inputsToSave: ["a", "b"], gradFunc: (e, t) => {
  let [n, s] = t, r = rt(n.shape, s.shape);
  return { a: () => {
    let o = e, u = At(n.shape, r);
    return u.length > 0 && (o = ve(o, u)), U(o, n.shape);
  }, b: () => {
    let o = e, u = At(s.shape, r);
    return u.length > 0 && (o = ve(o, u)), U(kt(o), s.shape);
  } };
} };
var CL = { kernelName: ci, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, r = s.shape.slice(), { axis: a } = n;
  ts(a, s.shape).forEach((l) => {
    r[l] = 1;
  });
  let o = U(e, r), u = V(o, Mn(s.shape, "float32"));
  return { x: () => u };
} };
var NL = { kernelName: Ho, inputsToSave: ["x"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => xe(e, ct(eb(n))) };
} };
var TL = { kernelName: fi, outputsToSave: [true], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => V(ge(we(1), ct(n)), e) };
} };
var $L = { kernelName: Cr, inputsToSave: ["x"], gradFunc: (e, t, n) => {
  let [s] = t, { reps: r } = n;
  return { x: () => {
    let i = je(s);
    if (s.rank === 1)
      for (let o = 0; o < r[0]; ++o)
        i = ie(i, qe(e, [o * s.shape[0]], [s.shape[0]]));
    else if (s.rank === 2)
      for (let o = 0; o < r[0]; ++o)
        for (let u = 0; u < r[1]; ++u)
          i = ie(i, qe(e, [o * s.shape[0], u * s.shape[1]], [s.shape[0], s.shape[1]]));
    else if (s.rank === 3)
      for (let o = 0; o < r[0]; ++o)
        for (let u = 0; u < r[1]; ++u)
          for (let l = 0; l < r[2]; ++l)
            i = ie(i, qe(e, [o * s.shape[0], u * s.shape[1], l * s.shape[2]], [s.shape[0], s.shape[1], s.shape[2]]));
    else if (s.rank === 4)
      for (let o = 0; o < r[0]; ++o)
        for (let u = 0; u < r[1]; ++u)
          for (let l = 0; l < r[2]; ++l)
            for (let c = 0; c < r[3]; ++c)
              i = ie(i, qe(e, [o * s.shape[0], u * s.shape[1], l * s.shape[2], c * s.shape[3]], [s.shape[0], s.shape[1], s.shape[2], s.shape[3]]));
    else
      throw new Error(`Gradient for tile operation is not implemented for rank-${s.rank} tensors yet.`);
    return i;
  } };
} };
var _L = { kernelName: mi, gradFunc: (e, t, n) => {
  let s = n, { perm: r } = s, a = nb(r);
  return { x: () => Ge(e, a) };
} };
var AL = { kernelName: Ko, gradFunc: (e, t, n) => {
  let s = n, { axis: r } = s;
  return { value: () => es(e, r) };
} };
var EL = { kernelName: mp, inputsToSave: ["segmentIds"], gradFunc: (e, t) => {
  let [n] = t;
  return { x: () => RL(e, n) };
} };
function RL(e, t) {
  let n = $r(t, je(t)), s = Zu(e, n), r = Zo(t, we(0, "int32")), a = s.rank - r.rank;
  for (let o = 0; o < a; ++o)
    r = Pn(r, o + 1);
  r = Ds(r, Mn(s.shape, "bool"));
  let i = je(s);
  return vn(r, s, i);
}
var DL = { kernelName: Xo, gradFunc: (e) => ({ x: () => je(e) }) };
var FL = [uI, Ez, Rz, Dz, Fz, Oz, Pz, zz, Mz, Lz, Bz, Vz, Gz, jz, Kz, Xz, Yz, Qz, Zz, Jz, eM, tM, sM, nM, iM, oM, uM, lM, cM, dM, tL, pM, hM, fM, mM, gM, yM, bM, vM, xM, wM, kM, SM, IM, CM, NM, TM, $M, _M, RM, Fx, Fx, DM, PM, LM, BM, VM, WM, UM, GM, HM, qM, jM, KM, XM, Ox, Ox, YM, QM, eL, nL, sL, rL, aL, iL, oL, uL, lL, cL, dL, pL, hL, fL, mL, gL, bL, yL, vL, Px, Px, zx, zx, xL, kL, wL, SL, IL, CL, NL, TL, $L, _L, AL, EL, DL];
for (let e of FL)
  V$(e);
var OL = {};
Ee(OL, { maxNorm: () => PL, minMaxNorm: () => LL, nonNeg: () => ML, unitNorm: () => zL });
function Bb(e, t) {
  return j(() => dn(ve(V(e, e), t, true)));
}
var ql = class extends re.Serializable {
  getConfig() {
    return {};
  }
};
var Vb = class extends ql {
  constructor(e) {
    super(), this.defaultMaxValue = 2, this.defaultAxis = 0, this.maxValue = e.maxValue != null ? e.maxValue : this.defaultMaxValue, this.axis = e.axis != null ? e.axis : this.defaultAxis;
  }
  apply(e) {
    return j(() => {
      let t = Bb(e, this.axis), n = Vn(t, 0, this.maxValue);
      return V(e, xe(n, ie(Dt(), t)));
    });
  }
  getConfig() {
    return { maxValue: this.maxValue, axis: this.axis };
  }
};
Vb.className = "MaxNorm";
re.registerClass(Vb);
var Wb = class extends ql {
  constructor(e) {
    super(), this.defaultAxis = 0, this.axis = e.axis != null ? e.axis : this.defaultAxis;
  }
  apply(e) {
    return j(() => xe(e, ie(Dt(), Bb(e, this.axis))));
  }
  getConfig() {
    return { axis: this.axis };
  }
};
Wb.className = "UnitNorm";
re.registerClass(Wb);
var Ub = class extends ql {
  apply(e) {
    return Xs(e);
  }
};
Ub.className = "NonNeg";
re.registerClass(Ub);
var Gb = class extends ql {
  constructor(e) {
    super(), this.defaultMinValue = 0, this.defaultMaxValue = 1, this.defaultRate = 1, this.defaultAxis = 0, this.minValue = e.minValue != null ? e.minValue : this.defaultMinValue, this.maxValue = e.maxValue != null ? e.maxValue : this.defaultMaxValue, this.rate = e.rate != null ? e.rate : this.defaultRate, this.axis = e.axis != null ? e.axis : this.defaultAxis;
  }
  apply(e) {
    return j(() => {
      let t = Bb(e, this.axis), n = ie(V(this.rate, Vn(t, this.minValue, this.maxValue)), V(1 - this.rate, t));
      return V(e, xe(n, ie(Dt(), t)));
    });
  }
  getConfig() {
    return { minValue: this.minValue, maxValue: this.maxValue, rate: this.rate, axis: this.axis };
  }
};
Gb.className = "MinMaxNorm";
re.registerClass(Gb);
var Mx = { maxNorm: "MaxNorm", minMaxNorm: "MinMaxNorm", nonNeg: "NonNeg", unitNorm: "UnitNorm" };
function Pt(e) {
  return _b(e);
}
function Lx(e, t = {}) {
  return Wl(e, re.SerializationMap.getMap().classNameMap, t, "constraint");
}
function zt(e) {
  if (e == null)
    return null;
  if (typeof e == "string") {
    let n = { className: e in Mx ? Mx[e] : e, config: {} };
    return Lx(n);
  } else
    return e instanceof ql ? e : Lx(e);
}
function PL(e) {
  return new Vb(e);
}
function zL(e) {
  return new Wb(e);
}
function ML() {
  return new Ub();
}
function LL(e) {
  return new Gb(e);
}
var BL = {};
Ee(BL, { constant: () => UL, glorotNormal: () => YL, glorotUniform: () => XL, heNormal: () => QL, heUniform: () => ZL, identity: () => jL, leCunNormal: () => JL, leCunUniform: () => eB, ones: () => WL, orthogonal: () => tB, randomNormal: () => HL, randomUniform: () => GL, truncatedNormal: () => qL, varianceScaling: () => KL, zeros: () => VL });
function VL() {
  return new Rb();
}
function WL() {
  return new Fp();
}
function UL(e) {
  return new Db(e);
}
function GL(e) {
  return new Fb(e);
}
function HL(e) {
  return new Ob(e);
}
function qL(e) {
  return new Pb(e);
}
function jL(e) {
  return new zb(e);
}
function KL(e) {
  return new xn(e);
}
function XL(e) {
  return new Op(e);
}
function YL(e) {
  return new Pp(e);
}
function QL(e) {
  return new zp(e);
}
function ZL(e) {
  return new Mp(e);
}
function JL(e) {
  return new Lp(e);
}
function eB(e) {
  return new Bp(e);
}
function tB(e) {
  return new Mb(e);
}
var nB = {};
Ee(nB, { Layer: () => He, RNN: () => Ar, RNNCell: () => Xl, activation: () => TV, add: () => PV, alphaDropout: () => vW, average: () => zV, averagePooling1d: () => Jy, averagePooling2d: () => ev, averagePooling3d: () => tv, avgPool1d: () => qV, avgPool2d: () => KV, avgPool3d: () => YV, avgPooling1d: () => jV, avgPooling2d: () => XV, avgPooling3d: () => QV, batchNormalization: () => UV, bidirectional: () => dW, concatenate: () => MV, conv1d: () => yV, conv2d: () => vV, conv2dTranspose: () => xV, conv3d: () => wV, conv3dTranspose: () => kV, convLstm2d: () => oW, convLstm2dCell: () => uW, cropping2D: () => IV, dense: () => $V, depthwiseConv2d: () => NV, dot: () => WV, dropout: () => _V, elu: () => pV, embedding: () => OV, flatten: () => EV, gaussianDropout: () => yW, gaussianNoise: () => bW, globalAveragePooling1d: () => ZV, globalAveragePooling2d: () => JV, globalMaxPool1d: () => hW, globalMaxPool2d: () => fW, globalMaxPooling1d: () => t0, globalMaxPooling2d: () => n0, gru: () => tW, gruCell: () => nW, input: () => ZB, inputLayer: () => dV, layerNormalization: () => GV, leakyReLU: () => fV, lstm: () => sW, lstmCell: () => rW, masking: () => xW, maxPool1d: () => mW, maxPool2d: () => gW, maxPooling1d: () => s0, maxPooling2d: () => r0, maxPooling3d: () => eW, maximum: () => LV, minimum: () => BV, multiply: () => VV, permute: () => FV, prelu: () => mV, reLU: () => hV, repeatVector: () => RV, reshape: () => DV, rnn: () => lW, separableConv2d: () => SV, simpleRNN: () => aW, simpleRNNCell: () => iW, softmax: () => gV, spatialDropout1d: () => AV, stackedRNNCells: () => cW, thresholdedReLU: () => bV, timeDistributed: () => pW, upSampling2d: () => CV, zeroPadding2d: () => HV });
async function sr(e) {
  if (e == null)
    return;
  let t = [], n = [], s = [];
  for (let r in e) {
    let a = e[r];
    if (typeof a != "number") {
      let i = a;
      t.push(i.data()), n.push(r), s.push(i);
    }
  }
  if (t.length > 0) {
    let r = await Promise.all(t);
    for (let a = 0; a < r.length; ++a)
      e[n[a]] = r[a][0];
    De(s);
  }
}
function cI(e) {
  if (e != null)
    for (let t in e) {
      let n = e[t];
      typeof n != "number" && n.dispose();
    }
}
var sB = 125;
var so = class {
  constructor() {
    this.validationData = null;
  }
  setParams(e) {
    this.params = e;
  }
  async onEpochBegin(e, t) {
  }
  async onEpochEnd(e, t) {
  }
  async onBatchBegin(e, t) {
  }
  async onBatchEnd(e, t) {
  }
  async onTrainBegin(e) {
  }
  async onTrainEnd(e) {
  }
  setModel(e) {
  }
};
var rB = class {
  constructor(e, t = 10) {
    e == null && (e = []), this.callbacks = e, this.queueLength = t;
  }
  append(e) {
    this.callbacks.push(e);
  }
  setParams(e) {
    for (let t of this.callbacks)
      t.setParams(e);
  }
  setModel(e) {
    for (let t of this.callbacks)
      t.setModel(e);
  }
  async onEpochBegin(e, t) {
    t == null && (t = {});
    for (let n of this.callbacks)
      await n.onEpochBegin(e, t);
  }
  async onEpochEnd(e, t) {
    t == null && (t = {});
    for (let n of this.callbacks)
      await n.onEpochEnd(e, t);
  }
  async onBatchBegin(e, t) {
    t == null && (t = {});
    for (let n of this.callbacks)
      await n.onBatchBegin(e, t);
  }
  async onBatchEnd(e, t) {
    t == null && (t = {});
    for (let n of this.callbacks)
      await n.onBatchEnd(e, t);
  }
  async onTrainBegin(e) {
    e == null && (e = {});
    for (let t of this.callbacks)
      await t.onTrainBegin(e);
  }
  async onTrainEnd(e) {
    e == null && (e = {});
    for (let t of this.callbacks)
      await t.onTrainEnd(e);
  }
};
var aB = class extends so {
  constructor() {
    super();
  }
  async onEpochBegin(e) {
    this.seen = 0, this.totals = {};
  }
  async onBatchEnd(e, t) {
    t == null && (t = {});
    let n = t.size == null ? 0 : t.size;
    this.seen += n;
    for (let s in t) {
      let r = t[s];
      if (typeof r == "number")
        this.totals.hasOwnProperty(s) || (this.totals[s] = 0), this.totals[s] = this.totals[s] + r * n;
      else {
        let a;
        s in this.totals ? a = this.totals[s] : this.totals[s] = 0;
        let i = j(() => ie(this.totals[s], V(r, n)));
        this.totals[s] = i, a != null && a.dispose();
      }
    }
  }
  async onEpochEnd(e, t) {
    if (t != null)
      for (let n of this.params.metrics)
        this.totals[n] != null && (typeof this.totals[n] == "number" ? t[n] = this.totals[n] / this.seen : j(() => {
          let s = V(xe(1, this.seen), this.totals[n]);
          t[n] = s, this.totals[n].dispose(), qt(t[n]);
        }));
  }
};
var iB = class extends so {
  async onTrainBegin(e) {
    this.epoch = [], this.history = {};
  }
  async onEpochEnd(e, t) {
    t == null && (t = {}), this.epoch.push(e);
    for (let n in t)
      this.history[n] == null && (this.history[n] = []), this.history[n].push(t[n]);
  }
  async syncData() {
    let e = [], t = [], n = [];
    for (let r in this.history) {
      let a = this.history[r];
      for (let i = 0; i < a.length; ++i)
        if (typeof a[i] != "number") {
          let o = a[i];
          e.push(o.data()), t.push(r), n.push(i);
        }
    }
    let s = await Promise.all(e);
    for (let r = 0; r < s.length; ++r)
      this.history[t[r]][n[r]].dispose(), this.history[t[r]][n[r]] = s[r][0];
  }
};
var oB = class extends so {
  constructor(e, t) {
    if (super(), this.currentEpoch = 0, this.nowFunc = e.nowFunc, this.nextFrameFunc = e.nextFrameFunc || HS, this.yieldEvery = t || "auto", this.yieldEvery === "auto" && (this.yieldEvery = sB), this.yieldEvery === "never" && e.onYield != null)
      throw new Error("yieldEvery is `never` but you provided an `onYield` callback. Either change `yieldEvery` or remove the callback");
    w.isNumber(this.yieldEvery) && (this.maybeWait = QP(this.maybeWait.bind(this), this.yieldEvery, this.nowFunc)), this.trainBegin = e.onTrainBegin, this.trainEnd = e.onTrainEnd, this.epochBegin = e.onEpochBegin, this.epochEnd = e.onEpochEnd, this.batchBegin = e.onBatchBegin, this.batchEnd = e.onBatchEnd, this.yield = e.onYield;
  }
  async maybeWait(e, t, n) {
    let s = [];
    this.yield != null && (await sr(n), s.push(this.yield(e, t, n))), s.push(this.nextFrameFunc()), await Promise.all(s);
  }
  async onEpochBegin(e, t) {
    this.currentEpoch = e, this.epochBegin != null && (await sr(t), await this.epochBegin(e, t));
  }
  async onEpochEnd(e, t) {
    let n = [];
    this.epochEnd != null && (await sr(t), n.push(this.epochEnd(e, t))), this.yieldEvery === "epoch" && n.push(this.nextFrameFunc()), await Promise.all(n);
  }
  async onBatchBegin(e, t) {
    this.batchBegin != null && (await sr(t), await this.batchBegin(e, t));
  }
  async onBatchEnd(e, t) {
    let n = [];
    this.batchEnd != null && (await sr(t), n.push(this.batchEnd(e, t))), this.yieldEvery === "batch" ? n.push(this.nextFrameFunc()) : w.isNumber(this.yieldEvery) && n.push(this.maybeWait(this.currentEpoch, e, t)), await Promise.all(n);
  }
  async onTrainBegin(e) {
    this.trainBegin != null && (await sr(e), await this.trainBegin(e));
  }
  async onTrainEnd(e) {
    this.trainEnd != null && (await sr(e), await this.trainEnd(e));
  }
};
function dI(e, t) {
  return e == null && (e = {}), e instanceof so ? [e] : Array.isArray(e) && e[0] instanceof so ? e : ht(e).map((s) => new oB(s, t));
}
var Ss = class {
  constructor() {
  }
  static registerCallbackConstructor(e, t) {
    w.assert(e >= 0 && Number.isInteger(e), () => `Verbosity level is expected to be an integer >= 0, but got ${e}`), Ss.checkForDuplicate(t), Ss.constructors[e] == null && (Ss.constructors[e] = []), Ss.constructors[e].push(t);
  }
  static checkForDuplicate(e) {
    for (let t in Ss.constructors)
      Ss.constructors[+t].forEach((s) => {
        if (s === e)
          throw new G("Duplicate callback constructor.");
      });
  }
  static clear() {
    Ss.constructors = {};
  }
  static createCallbacks(e) {
    let t = [];
    for (let n in Ss.constructors) {
      let s = +n;
      e >= s && t.push(...Ss.constructors[s]);
    }
    return t.map((n) => new n());
  }
};
var Hb = Ss;
Hb.constructors = {};
function pI(e, t, n, s, r, a, i, o, u) {
  let l = new iB(), c = [new aB(), ...Hb.createCallbacks(t)];
  e != null && c.push(...e), c.push(l);
  let p = new rB(c);
  return p.setParams({ epochs: n, initialEpoch: s, samples: r, steps: a, batchSize: i, verbose: t, doValidation: o, metrics: u }), { callbackList: p, history: l };
}
function gs(e, t = {}, n = false) {
  return Wl(e, re.SerializationMap.getMap().classNameMap, t, "layer", n);
}
function Rd(e, t) {
  return j(() => {
    e.dtype !== "float32" && (e = le(e, "float32"));
    let n = ve(Gl(e), t, true), s = Ll(n.shape, Dt()), r = dn($r(n, s));
    return xe(e, r);
  });
}
function vi(e, t) {
  return j(() => It(Gl(ge(t, e)), -1));
}
function Wp(e, t) {
  return j(() => It(Lt(ge(t, e)), -1));
}
function nu(e, t) {
  return j(() => {
    let n = ge(e, t), s = Vn(Lt(e), Dt(), Number.MAX_VALUE), r = Lt(xe(n, s));
    return V(100, It(r, -1));
  });
}
function uB(e, t) {
  return j(() => {
    let n = Vn(t, Dt(), Number.MAX_VALUE), s = Qn(ie(1, n)), r = Vn(e, Dt(), Number.MAX_VALUE), a = Qn(ie(1, r));
    return It(Gl(ge(s, a)), -1);
  });
}
function lB(e, t) {
  return j(() => {
    let n = $r(0, ge(1, V(e, t)));
    return It(Gl(n), -1);
  });
}
function cB(e, t) {
  return j(() => {
    let n = $r(0, ge(1, V(e, t)));
    return It(n, -1);
  });
}
function dB(e, t) {
  return j(() => {
    let n = ve(V(e, t), -1), s = As(V(ge(1, e), t), -1);
    return $r(0, ie(1, ge(s, n)));
  });
}
function pB(e, t) {
  return j(() => {
    let n = Math.log(2), s = ge(t, e), r = ge(ie(s, Bl(V(-2, s))), n);
    return It(r, -1);
  });
}
function tl(e, t, n = false) {
  return j(() => {
    if (n)
      t = gb(t);
    else {
      let s = ve(t, t.shape.length - 1, true);
      t = xe(t, s);
    }
    return t = Vn(t, Dt(), 1 - Dt()), kt(ve(V(le(e, "float32"), Qn(t)), t.shape.length - 1));
  });
}
function Dd(e, t, n = false) {
  return j(() => {
    let s = le(kp(lz(e)), "int32");
    t = Vn(t, Dt(), 1 - Dt());
    let r = t.shape, a = U(Sd(s, r[r.length - 1]), r);
    return tl(a, t, n);
  });
}
function hB(e, t) {
  if (!w.arraysEqual(e.shape, t.shape))
    throw new G(`logits and labels must have the same shape, but got shapes ${JSON.stringify(e.shape)} and ${JSON.stringify(t.shape)}`);
  return j(() => {
    let n = Xs(t), s = kt(Lt(t));
    return ie(ge(n, V(t, e)), ib(Yn(s)));
  });
}
function Up(e, t) {
  return j(() => {
    let n;
    return n = Vn(t, Dt(), 1 - Dt()), n = Qn(xe(n, ge(1, n))), It(hB(e, n), -1);
  });
}
function fB(e, t) {
  return j(() => {
    let n = Vn(e, Dt(), 1), s = Vn(t, Dt(), 1);
    return ve(V(e, Qn(xe(n, s))), -1);
  });
}
function mB(e, t) {
  return j(() => {
    let n = Qn(ie(Dt(), t));
    return It(ge(t, V(e, n)), -1);
  });
}
function qb(e, t) {
  return j(() => {
    let n = Rd(e, -1), s = Rd(t, -1), r = V(n, s);
    return kt(ve(r, -1));
  });
}
var Fd = { meanSquaredError: vi, meanAbsoluteError: Wp, meanAbsolutePercentageError: nu, meanSquaredLogarithmicError: uB, squaredHinge: lB, hinge: cB, categoricalHinge: dB, logcosh: pB, categoricalCrossentropy: tl, sparseCategoricalCrossentropy: Dd, binaryCrossentropy: Up, kullbackLeiblerDivergence: fB, poisson: mB, cosineProximity: qb };
function Xf(e) {
  if (typeof e == "string") {
    if (e in Fd)
      return Fd[e];
    let t = `Unknown loss ${e}`;
    throw e.toLowerCase().includes("softmaxcrossentropy") && (t = `Unknown loss ${e}. Use "categoricalCrossentropy" as the string name for tf.losses.softmaxCrossEntropy`), new G(t);
  } else
    return e;
}
function jb(e, t) {
  return j(() => {
    let n = V(0.5, Zn(t)), s = Rp(Un(t, n), e.dtype);
    return It(Xn(e, s), -1);
  });
}
function Kb(e, t) {
  return j(() => Rp(Xn(Xu(e, -1), Xu(t, -1)), "float32"));
}
function hI(e, t) {
  return j(() => le(ve(Ds(Xn(e, 1), Xn(t, 1))), "float32"));
}
function gB(e, t) {
  return j(() => le(ve(Ds(Xn(e, 1), Xn(t, 0))), "float32"));
}
function bB(e, t) {
  return j(() => le(ve(Ds(Xn(e, 0), Xn(t, 1))), "float32"));
}
function fI(e, t) {
  return j(() => {
    let n = hI(e, t), s = bB(e, t), r = ie(n, s);
    return le(vn(Un(r, 0), xe(n, r), 0), "float32");
  });
}
function yB(e, t) {
  return j(() => {
    let n = hI(e, t), s = gB(e, t), r = ie(n, s);
    return le(vn(Un(r, 0), xe(n, r), 0), "float32");
  });
}
function mI(e, t) {
  return Up(e, t);
}
function gI(e, t) {
  return e.rank === t.rank && (e = mr(e, [e.rank - 1])), t = Xu(t, -1), t.dtype !== e.dtype && (t = le(t, e.dtype)), le(Xn(e, t), "float32");
}
var vB = vi;
var xB = vi;
var wB = Wp;
var kB = Wp;
var SB = nu;
var IB = nu;
var Xb = tl;
var CB = qb;
var bI = Dd;
var Od = { binaryAccuracy: jb, categoricalAccuracy: Kb, precision: fI, categoricalCrossentropy: Xb, sparseCategoricalCrossentropy: bI, mse: vB, MSE: xB, mae: wB, MAE: kB, mape: SB, MAPE: IB, cosine: CB };
function NB(e) {
  if (typeof e == "string" && e in Od)
    return Od[e];
  if (typeof e != "string" && e != null)
    return e;
  throw new G(`Unknown metric ${e}`);
}
function Xc(e) {
  if (Cs(e !== null, `Unknown LossOrMetricFn ${e}`), typeof e == "string")
    return e;
  {
    let t;
    for (let n of Object.keys(Fd))
      if (Fd[n] === e) {
        t = n;
        break;
      }
    if (t !== void 0)
      return t;
    for (let n of Object.keys(Od))
      if (Od[n] === e) {
        t = n;
        break;
      }
    return t !== void 0 ? t : e.name;
  }
}
function TB(e) {
  let t = { Adagrad: () => Li.adagrad(0.01), Adadelta: () => Li.adadelta(1, 0.95, Dt()), Adam: () => Li.adam(1e-3, 0.9, 0.999, Dt()), Adamax: () => Li.adamax(2e-3, 0.9, 0.999, Dt(), 0), RMSProp: () => Li.rmsprop(1e-3, 0.9, 0, Dt()), SGD: () => Li.sgd(0.01) };
  if (t.adagrad = t.Adagrad, t.adadelta = t.Adadelta, t.adam = t.Adam, t.adamax = t.Adamax, t.rmsprop = t.RMSProp, t.sgd = t.SGD, e in t)
    return t[e]();
  throw new G(`Unknown Optimizer ${e}`);
}
var Bx = 1 * 1024 * 1024;
function Vx(e, t, n = false) {
  if (e == null || typeof e != "object" || Object.getPrototypeOf(e) !== Object.prototype || !Tm(e))
    throw new Error("User-defined metadata is expected to be a JSON object, but is not.");
  if (n) {
    let s = JSON.stringify(e);
    s.length > Bx && console.warn(`User-defined metadata of model "${t}" is too large in size (length=${s.length} when serialized). It is not recommended to store such large objects in user-defined metadata. Please make sure its serialized length is <= ${Bx}.`);
  }
}
function Tm(e) {
  if (e === null)
    return true;
  if (typeof e == "object")
    if (Object.getPrototypeOf(e) === Object.prototype) {
      let t = Object.keys(e);
      for (let n of t)
        if (typeof n != "string" || !Tm(e[n]))
          return false;
      return true;
    } else if (Array.isArray(e)) {
      for (let t of e)
        if (!Tm(t))
          return false;
      return true;
    } else
      return false;
  else {
    let t = typeof e;
    return t === "string" || t === "number" || t === "boolean";
  }
}
function $B(e, t, n, s = console.log) {
  let r = AB(e), a = ["Layer (type)", "Input Shape", "Output shape", "Param #"];
  r ? (t = t || 90, n = n || [0.32, 0.61, 0.89, 1]) : (t = t || 115, n = n || [0.24, 0.48, 0.7, 0.8, 1]), n[n.length - 1] <= 1 && (n = n.map((c) => Math.floor(t * c)));
  let i;
  if (!r) {
    a.push("Receives inputs"), i = [];
    for (let c in e.nodesByDepth)
      i.push(...e.nodesByDepth[c]);
  }
  s("_".repeat(t)), Pd(a, n, s), s("=".repeat(t));
  let o = e.layers;
  for (let c = 0; c < o.length; ++c)
    r ? EB(o[c], n, s) : RB(o[c], n, i, s), s((c === o.length - 1 ? "=" : "_").repeat(t));
  e.checkTrainableWeightsConsistency();
  let u = _B(e), l = _d(e.nonTrainableWeights);
  s(`Total params: ${u + l}`), s(`Trainable params: ${u}`), s(`Non-trainable params: ${l}`), s("_".repeat(t));
}
function _B(e) {
  let t;
  return e.collectedTrainableWeights != null ? t = _d(e.collectedTrainableWeights) : t = _d(e.trainableWeights), t;
}
function AB(e) {
  let t = true, n = [], s = [];
  for (let r in e.nodesByDepth)
    n.push(e.nodesByDepth[r]);
  for (let r of n) {
    if (r.length > 1 || r.length === 1 && r[0].inboundLayers.length > 1) {
      t = false;
      break;
    }
    s.push(...r);
  }
  if (t)
    for (let r of e.layers) {
      let a = false;
      for (let i of r.inboundNodes)
        if (s.indexOf(i) !== -1)
          if (a) {
            t = false;
            break;
          } else
            a = true;
      if (!t)
        break;
    }
  return t;
}
function Pd(e, t, n = console.log) {
  let s = "";
  for (let r = 0; r < e.length; ++r)
    r > 0 && (s = s.slice(0, s.length - 1) + " "), s += e[r], s = s.slice(0, t[r]), s += " ".repeat(t[r] - s.length);
  n(s);
}
function EB(e, t, n) {
  let s, r;
  try {
    r = e.inboundNodes.map((u) => JSON.stringify(u.inputShapes)).join(",");
  } catch (u) {
    r = "multiple";
  }
  try {
    s = JSON.stringify(e.outputShape);
  } catch (u) {
    s = "multiple";
  }
  let a = e.name, i = e.getClassName(), o = [`${a} (${i})`, r, s, e.countParams().toString()];
  Pd(o, t, n);
}
function RB(e, t, n, s) {
  let r, a;
  try {
    a = e.inboundNodes.map((p) => JSON.stringify(p.inputShapes)).join(",");
  } catch (p) {
    a = "multiple";
  }
  try {
    r = JSON.stringify(e.outputShape);
  } catch (p) {
    r = "multiple";
  }
  let i = [];
  for (let p of e.inboundNodes)
    if (!(n != null && n.length > 0 && n.indexOf(p) === -1))
      for (let d = 0; d < p.inboundLayers.length; ++d) {
        let h = p.inboundLayers[d].name, f = p.nodeIndices[d], m = p.tensorIndices[d];
        i.push(`${h}[${f}][${m}]`);
      }
  let o = e.name, u = e.getClassName(), l = i.length === 0 ? "" : i[0], c = [`${o} (${u})`, a, r, e.countParams().toString(), l];
  Pd(c, t, s);
  for (let p = 1; p < i.length; ++p)
    Pd(["", "", "", "", i[p]], t, s);
}
function yI(e, t, n) {
  return (e === "inboundNodes" || e === "outputLayers" || e === "inputLayers") && t === 0 && typeof n == "string";
}
function nl(e, t) {
  if (e === null)
    return null;
  if (typeof e == "string")
    return Xr(e);
  if (typeof e == "number" || typeof e == "boolean")
    return e;
  if (e instanceof Array) {
    let n = [], s = e.length;
    for (let r = 0; r < s; ++r) {
      let a = e[r];
      yI(t, r, a) ? n.push(a) : n.push(nl(a, t));
    }
    return n;
  } else {
    let n = {};
    for (let s of Object.keys(e)) {
      let r = e[s];
      if (s === "name" && typeof r == "string")
        n[s] = r;
      else {
        let a = Xr(s);
        n[a] = nl(r, a);
      }
    }
    return n;
  }
}
function $m(e, t) {
  if (e == null)
    return null;
  if (typeof e == "string")
    return Vs(e);
  if (typeof e == "number" || typeof e == "boolean")
    return e;
  if (e instanceof Array) {
    let n = [], s = e.length;
    for (let r = 0; r < s; ++r) {
      let a = e[r];
      yI(t, r, a) ? n.push(a) : n.push($m(a, t));
    }
    return n;
  } else {
    let n = {};
    for (let s of Object.keys(e)) {
      let r = e[s], a = Vs(s);
      (s === "name" || s === "className") && typeof r == "string" ? n[a] = r : n[a] = $m(r, s);
    }
    return n;
  }
}
var vI = "0.0.0";
var Is = class extends He {
  constructor(e) {
    if (super({}), this.containerNodes = /* @__PURE__ */ new Set(), this.name = e.name, this.name == null) {
      let b = this.getClassName().toLowerCase();
      this.name = Ep(b);
    }
    if (this.supportsMasking = false, this.trainable_ = true, Array.isArray(e.inputs) ? this.inputs = e.inputs.slice() : this.inputs = [e.inputs], Array.isArray(e.outputs) ? this.outputs = e.outputs.slice() : this.outputs = [e.outputs], lr(this.inputs).length !== this.inputs.length)
      throw new G(`The list of inputs passed to the model is redundant. All inputs should only appear once. Found: ${this.inputs.map((b) => b.name)}`);
    lr(this.outputs).length !== this.outputs.length && console.warn(`The list of outputs passed to the model is redundant. All outputs should only appear once. Found: ${this.outputs.map((b) => b.name)}`), this.inputLayers = [], this.inputLayersNodeIndices = [], this.inputLayersTensorIndices = [], this.outputLayers = [], this.outputLayersNodeIndices = [], this.outputLayersTensorIndices = [], this.layers = [], this.internalContainerRefs = [];
    for (let b of this.outputs) {
      let y = b.sourceLayer, v = b.nodeIndex, x = b.tensorIndex;
      this.outputLayers.push(y), this.outputLayersNodeIndices.push(v), this.outputLayersTensorIndices.push(x);
    }
    for (let b of this.inputs) {
      let y = b.sourceLayer, v = b.nodeIndex, x = b.tensorIndex;
      Cs(v === 0, "input layer has >1 nodes"), Cs(x === 0, "input layer has >1 tensors"), this.inputLayers.push(y), this.inputLayersNodeIndices.push(v), this.inputLayersTensorIndices.push(x);
    }
    this.inputNames = [], this.outputNames = [], this.feedInputShapes = [], this.feedInputNames = [], this.feedOutputNames = [];
    for (let b = 0; b < this.inputLayers.length; b++) {
      let y = this.inputLayers[b];
      if (!(y instanceof tu))
        throw new TypeError(`Input layers to a LayersModel must be InputLayer objects. Received inputs: ${e.inputs}. Input ${b} (0-based) originates from layer type ${y.getClassName()}.`);
      this.inputNames.push(y.name), this.feedInputShapes.push(y.batchInputShape), this.feedInputNames.push(y.name);
    }
    for (let b of this.outputLayers)
      this.outputNames.push(b.name);
    this.internalInputShapes = this.inputs.map((b) => b.shape), this.internalOutputShapes = this.outputs.map((b) => b.shape);
    let t = {}, n = {}, s = {}, r = {}, a = {}, i = [], o = (b, y, v, x, k, I) => {
      (x == null || k == null || I == null) && (x = b.sourceLayer, k = b.nodeIndex, I = b.tensorIndex);
      let $ = x.inboundNodes[k];
      if (v.indexOf($) !== -1)
        throw new fs(`The tensor ${b.name} at layer "${x.name}" is part of a cycle.`);
      if (y.indexOf($) !== -1)
        return;
      this.containerNodes.add(Is.nodeKey(x, k)), x.id in a || (a[x.id] = Object.keys(a).length), v.indexOf($) === -1 && v.push($);
      let R = $.inboundLayers.length;
      for (let E = 0; E < R; E++) {
        let P = $.inputTensors[E], A = $.inboundLayers[E], O = $.nodeIndices[E], T = $.tensorIndices[E];
        o(P, y, v, A, O, T);
      }
      for (y.push($); v.indexOf($) >= 0; )
        v.splice(v.indexOf($), 1);
      i.push($);
    }, u = [], l = [];
    for (let b of this.outputs)
      o(b, u, l);
    let c = i.slice().reverse();
    for (let b of c) {
      n[b.id] = b, b.id in t || (t[b.id] = 0);
      let y = t[b.id], v = s[b.outboundLayer.id] == null ? 0 : s[b.outboundLayer.id];
      y = Math.max(y, v), s[b.outboundLayer.id] = y, r[b.outboundLayer.id] = b.outboundLayer, t[b.id] = y;
      for (let x = 0; x < b.inboundLayers.length; x++) {
        let k = b.inboundLayers[x], I = b.nodeIndices[x], $ = k.inboundNodes[I], R = t[$.id] == null ? 0 : t[$.id];
        t[$.id] = Math.max(y + 1, R), n[$.id] = $;
      }
    }
    let p = {};
    for (let b in t) {
      let y = t[b];
      y in p || (p[y] = []), p[y].push(n[b]);
    }
    let d = {};
    for (let b in s) {
      let y = s[b];
      y in d || (d[y] = []), d[y].push(r[b]);
    }
    let h = Object.keys(d).map((b) => parseInt(b, 10)).sort(qc);
    this.layers = [];
    for (let b of h) {
      let y = d[b];
      y.sort((v, x) => {
        let k = a[v.id], I = a[x.id];
        return k < I ? -1 : k > I ? 1 : 0;
      });
      for (let v of y)
        v instanceof Is && this.internalContainerRefs.push(v), this.layers.push(v);
    }
    this.layersByDepth = d, h = Object.keys(p).map((b) => parseInt(b, 10)).sort(qc);
    let f = this.inputs.slice(), m = [];
    for (let b of h)
      for (let y of p[b]) {
        let v = y.outboundLayer;
        if (v != null) {
          for (let x of y.inputTensors)
            if (f.indexOf(x) === -1)
              throw new fs(`Graph disconnected: cannot obtain value for tensor ${x} at layer "${v.name}". The following previous layers were accessed without issue: ${m}`);
          for (let x of y.outputTensors)
            f.push(x);
          m.push(v.name);
        }
      }
    this.nodesByDepth = p;
    let g = this.layers.map((b) => b.name);
    for (let b of g) {
      let y = g.filter((v) => v === b).length;
      if (y !== 1)
        throw new fs(`The name "${b}" is used ${y} times in the model. All layer names should be unique. Layer names: ` + JSON.stringify(g));
    }
    this.outboundNodes = [], this.inboundNodes = [], new Vp({ outboundLayer: this, inboundLayers: [], nodeIndices: [], tensorIndices: [], inputTensors: this.inputs, outputTensors: this.outputs, inputMasks: this.inputs.map((b) => null), outputMasks: this.outputs.map((b) => null), inputShapes: this.inputs.map((b) => b.shape), outputShapes: this.outputs.map((b) => b.shape) }), this.built = true, this._refCount = 1;
  }
  assertNotDisposed() {
    if (this._refCount === 0)
      throw new Error(`Container '${this.name}' is already disposed.`);
  }
  dispose() {
    this.assertNotDisposed();
    let e = { refCountAfterDispose: null, numDisposedVariables: 0 };
    if (--this._refCount === 0) {
      for (let t of this.layers)
        e.numDisposedVariables += t.dispose().numDisposedVariables;
      for (let t of this.internalContainerRefs)
        e.numDisposedVariables += t.dispose().numDisposedVariables;
    }
    return e.refCountAfterDispose = this._refCount, e;
  }
  get trainable() {
    return this.trainable_;
  }
  set trainable(e) {
    this.layers.forEach((t) => {
      t._trainableWeights.forEach((n) => n.trainable = e);
    }), this.trainable_ = e;
  }
  get trainableWeights() {
    if (this._trainableWeights.length > 0)
      throw new G("Container instance unexpectedly contains _trainableWeights.The trainable weights of a Container are a union of the trainable weights of its consituent Layers. Its own _trainableWeights must remain an empty Array.");
    if (!this.trainable)
      return [];
    let e = [];
    for (let t of this.layers)
      e = e.concat(t.trainableWeights);
    return e;
  }
  get nonTrainableWeights() {
    let e = [];
    for (let t of this.layers)
      e.push(...t.nonTrainableWeights);
    if (!this.trainable) {
      let t = [];
      for (let n of this.layers)
        t.push(...n.trainableWeights);
      return t.concat(e);
    }
    return e;
  }
  get weights() {
    return this.trainableWeights.concat(this.nonTrainableWeights);
  }
  loadWeights(e, t = true) {
    let n = {}, s = 0;
    for (let a of this.layers)
      for (let i of a.weights) {
        if (n[i.originalName] != null)
          throw new G(`Duplicate weight name: ${i.originalName}`);
        n[i.originalName] = i, s++;
      }
    let r = [];
    for (let a in e) {
      let i = a;
      if (n[a] == null) {
        let o = a.split("/");
        i = o.slice(0, -2).concat([o[o.length - 1]]).join("/");
      }
      if (n[i] != null)
        r.push([n[i], e[a]]);
      else if (t)
        throw new G(`Provided weight data has no target variable: ${a}`);
      delete n[i];
    }
    if (t) {
      let a = [];
      for (let i in n)
        a.push(i);
      if (a.length > 0)
        throw new G(`${a.length} of ${s} weights are not set: ${a}`);
    }
    Lb(r);
  }
  updatedConfig() {
    let e = this.getConfig(), t = {};
    return t.className = this.getClassName(), t.config = e, t.kerasVersion = `tfjs-layers ${vI}`, t.backend = "TensorFlow.js", t;
  }
  toJSON(e, t = true) {
    let n = $m(this.updatedConfig());
    return t ? JSON.stringify(n) : n;
  }
  call(e, t) {
    return j(() => {
      e = ht(e);
      let n = new Zr();
      for (let s = 0; s < this.inputs.length; ++s)
        n.add(this.inputs[s], e[s]);
      return Fu(this.outputs, n, t);
    });
  }
  computeMask(e, t) {
    return j(() => {
      e = ht(e);
      let n;
      return t == null ? n = ma(null, e.length) : n = ht(t), this.runInternalGraph(e, n)[1];
    });
  }
  computeOutputShape(e) {
    let t = $d(e);
    if (t.length !== this.inputLayers.length)
      throw new G(`Invalid inputShape argument ${e}: model has ${this.inputLayers.length} tensor inputs.`);
    let n = {};
    for (let i = 0; i < t.length; i++) {
      let o = this.inputLayers[i], u = t[i], l = o.name + "_0_0";
      n[l] = u;
    }
    let s = Object.keys(this.nodesByDepth).map((i) => parseInt(i, 10)).sort(qc);
    if (s.length > 1)
      for (let i of s) {
        let o = this.nodesByDepth[i];
        for (let u of o) {
          let l = u.outboundLayer;
          if (this.inputLayers.map((f) => f.id).indexOf(l.id) !== -1)
            continue;
          let c = [];
          for (let f = 0; f < u.inboundLayers.length; f++) {
            let m = u.inboundLayers[f], g = u.nodeIndices[f], b = u.tensorIndices[f], y = `${m.name}_${g}_${b}`, v = n[y];
            c.push(v);
          }
          let p = l.computeOutputShape(bn(c)), d = $d(p), h = l.inboundNodes.indexOf(u);
          for (let f = 0; f < d.length; f++) {
            let m = `${l.name}_${h}_${f}`;
            n[m] = d[f];
          }
        }
      }
    let r = [], a = [];
    for (let i = 0; i < this.outputLayers.length; i++) {
      let o = this.outputLayers[i], u = this.outputLayersNodeIndices[i], l = this.outputLayersTensorIndices[i], c = `${o.name}_${u}_${l}`;
      a.push(c);
    }
    for (let i = 0; i < a.length; i++) {
      let o = a[i];
      Cs(o in n), r.push(n[o]);
    }
    return bn(r);
  }
  runInternalGraph(e, t) {
    t == null && (t = ma(null, e.length));
    let n = {};
    for (let o = 0; o < this.inputs.length; ++o) {
      let u = this.inputs[o], l = e[o], c = t[o];
      n[u.id] = [l, c];
    }
    let s = Object.keys(this.nodesByDepth).map((o) => parseInt(o, 10)).sort(qc);
    for (let o of s) {
      let u = this.nodesByDepth[o];
      for (let l of u) {
        let c = l.outboundLayer, p = l.inputTensors, d = l.outputTensors, h = new Array();
        for (let f of p)
          f.id in n && h.push(n[f.id]);
        if (h.length === p.length) {
          let f = {}, m, g, b, y;
          if (l.callArgs != null && (f = l.callArgs), h.length === 1) {
            let [v, x] = h[0];
            f.mask == null && (f.mask = x), b = ht(c.call(v, f)), y = ht(c.computeMask(v, x)), m = [v], g = [x];
          } else
            m = h.map((v) => v[0]), g = h.map((v) => v[1]), f.mask == null && (f.mask = g), b = ht(c.call(m, f)), y = ht(c.computeMask(m, g));
          if (c.activityRegularizer)
            throw new Fe("LayersModel invocation with concrete Tensor value(s) in the presence of activity regularizer(s) is not supported yet.");
          for (let v = 0; v < d.length; ++v) {
            let x = d[v], k = b[v], I = y[v];
            n[x.id] = [k, I];
          }
        }
      }
    }
    let r = [], a = [], i = [];
    for (let o of this.outputs) {
      Cs(o.id in n, `Could not compute output ${o.name} : ${o.id}`);
      let [u, l] = n[o.id];
      i.push(u.shape), r.push(u), a.push(l);
    }
    return [r, a, i];
  }
  buildNodeConversionMap(e) {
    let t = {}, n;
    for (let s of this.layers) {
      n = s instanceof Is ? 1 : 0;
      for (let r = 0; r < s.inboundNodes.length; r++) {
        let a = Is.nodeKey(s, r);
        this.containerNodes.has(a) && (t[a] = n, n += 1);
      }
    }
    return t;
  }
  getLayer(e, t) {
    if (t != null) {
      if (this.layers.length <= t)
        throw new G(`Was asked to retrieve layer at index ${t}, but model only has ${this.layers.length} layer(s).`);
      return this.layers[t];
    } else if (e == null)
      throw new G("Provide either a layer name or layer index");
    for (let n of this.layers)
      if (n.name === e)
        return n;
    throw new G(`No such layer: ${e}`);
  }
  calculateLosses() {
    return j(() => {
      let e = [];
      for (let t of this.layers)
        for (let n = 0; n < t.inboundNodes.length; ++n) {
          let s = Is.nodeKey(t, n);
          this.containerNodes.has(s) && e.push(...t.calculateLosses());
        }
      return e;
    });
  }
  getConfig() {
    let e = { name: this.name }, t = this.buildNodeConversionMap(this.layers), n = [];
    for (let a of this.layers) {
      let i = a.getClassName(), o = a.getConfig(), u = [];
      for (let c = 0; c < a.inboundNodes.length; c++) {
        let p = a.inboundNodes[c], d = Is.nodeKey(a, c), h = {};
        if (this.containerNodes.has(d)) {
          if (p.callArgs)
            try {
              JSON.stringify(p.callArgs), h = p.callArgs;
            } catch (f) {
              console.warn(`Layer ${a.name} was passed non-serializable keyword arguments: ${p.callArgs}. They will not be included in the serialized model (and thus will be missing at deserialization time).`), h = {};
            }
          if (p.inboundLayers.length > 0) {
            let f = [];
            for (let m = 0; m < p.inboundLayers.length; m++) {
              let g = p.inboundLayers[m], b = p.nodeIndices[m], y = p.tensorIndices[m], v = Is.nodeKey(g, b), x = t[v];
              x == null && (x = 0), f.push([g.name, x, y, h]);
            }
            u.push(f);
          }
        }
      }
      let l = {};
      l.name = a.name, l.className = i, l.config = o, l.inboundNodes = u, n.push(l);
    }
    e.layers = n;
    let s = [];
    for (let a = 0; a < this.inputLayers.length; a++) {
      let i = this.inputLayers[a], o = this.inputLayersNodeIndices[a], u = Is.nodeKey(i, o);
      if (!this.containerNodes.has(u))
        continue;
      let l = t[u];
      l == null && (l = 0);
      let c = this.inputLayersTensorIndices[a];
      s.push([i.name, l, c]);
    }
    e.inputLayers = s;
    let r = [];
    for (let a = 0; a < this.outputLayers.length; a++) {
      let i = this.outputLayers[a], o = this.outputLayersNodeIndices[a], u = Is.nodeKey(i, o);
      if (!this.containerNodes.has(u))
        continue;
      let l = t[u];
      l == null && (l = 0);
      let c = this.outputLayersTensorIndices[a];
      r.push([i.name, l, c]);
    }
    return e.outputLayers = r, e;
  }
  static fromConfig(e, t, n = {}, s = false) {
    let r = {}, a = {};
    function i(m, g) {
      m.name in a ? a[m.name].push(g) : a[m.name] = [g];
    }
    function o(m, g) {
      let b = [], y;
      for (let v of g) {
        let x = v[0], k = v[1], I = v[2];
        if (y = v[3] == null ? {} : v[3], !(x in r)) {
          i(m, g);
          return;
        }
        let $ = r[x];
        if ($.inboundNodes.length <= k) {
          i(m, g);
          return;
        }
        let R = $.inboundNodes[k];
        b.push(R.outputTensors[I]);
      }
      b.length > 0 && m.apply(bn(b), y);
    }
    function u(m) {
      let g = m.name, b = gs(m, t.customObjects != null ? t.customObjects : {});
      b.setFastWeightInitDuringBuild(s), r[g] = b, m.inboundNodes.forEach((v) => {
        if (!(v instanceof Array))
          throw new G(`Corrupted configuration, expected array for nodeData: ${v}`);
        i(b, v);
      });
    }
    let l = t.name, c = t.layers;
    for (let m of c)
      u(m);
    for (; !YP(a); )
      for (let m of c) {
        let g = r[m.name];
        if (g.name in a) {
          let b = a[g.name];
          delete a[g.name];
          for (let y of b)
            o(g, y);
        }
      }
    let p = [], d = [], h = t.inputLayers;
    for (let m of h) {
      let g = m[0], b = m[1], y = m[2];
      Cs(g in r);
      let x = r[g].inboundNodes[b].outputTensors;
      p.push(x[y]);
    }
    let f = t.outputLayers;
    for (let m of f) {
      let g = m[0], b = m[1], y = m[2];
      Cs(g in r);
      let x = r[g].inboundNodes[b].outputTensors;
      d.push(x[y]);
    }
    return new e({ inputs: p, outputs: d, name: l });
  }
  get stateful() {
    if (this._stateful)
      throw new G("Container instance unexpectedly has _stateful = true. The statefulness of a Container is determined by the Layers it contains. Its _stateful property must remain the default false.");
    for (let e of this.layers)
      if (e.stateful)
        return true;
    return false;
  }
  resetStates() {
    j(() => {
      this.layers.forEach((e) => {
        e.stateful && e.resetStates();
      });
    });
  }
};
function DB(e, t, n) {
  let s = t.length;
  if (e == null || Array.isArray(e) && e.length === 0)
    return t.map((r) => null);
  if (s === 1)
    return Array.isArray(e) && e.length === 1 ? e : typeof e == "object" && t[0] in e ? [e[t[0]]] : [e];
  if (Array.isArray(e)) {
    if (e.length !== s)
      throw new Error(`Provided ${n} is an array of ${e.length} element(s), but the model has ${s} outputs. Make sure a set of weights is provided for each model output.`);
    return e;
  } else if (typeof e == "object" && Object.keys(e).length > 0 && typeof e[Object.keys(e)[0]] == "object") {
    let r = [];
    return t.forEach((a) => {
      a in e ? r.push(e[a]) : r.push(null);
    }), r;
  } else
    throw new Error(`The model has multiple (${s}) outputs, so ${n} must be either an array with ${s} elements or an object with ${t} keys. Provided ${n} not understood: ${JSON.stringify(e)}`);
}
function xI(e, t) {
  return DB(e, t, "classWeight");
}
async function wI(e, t, n, s) {
  if (t != null || s != null)
    throw new Error("Support sampleWeight is not implemented yet");
  if (n != null) {
    let r = j(() => {
      if (e.shape.length === 1)
        return ur(e);
      if (e.shape.length === 2) {
        if (e.shape[1] > 1)
          return Xu(e, 1);
        if (e.shape[1] === 1)
          return U(e, [e.shape[0]]);
        throw new Error(`Encountered unexpected last-dimension size (${e.shape[1]}) during handling of class weights. The size is expected to be >= 1.`);
      } else
        throw new Error(`Unexpected rank of target (y) tensor (${e.rank}) during handling of class weights. The rank is expected to be 1 or 2.`);
    }), a = Array.from(await r.data());
    De(r);
    let i = [];
    return a.forEach((o) => {
      if (n[o] == null)
        throw new Error(`classWeight must contain all classes in the training data. The class ${o} exists in the data but not in classWeight`);
      i.push(n[o]);
    }), Zt(i, "float32");
  } else
    return null;
}
function FB(e, t) {
  return V(e, t);
}
var OB = 32;
function kI(e, t) {
  let n, s, r = t;
  n = r.xs, s = r.ys, w.assert(n != null && s != null, () => `A Dataset iterator for fitDataset() is expected to generate objects of the form \`{xs: xVal, ys: yVal}\`, where the two values may be \`tf.Tensor\`, an array of Tensors, or a map of string to Tensor.  The provided Dataset instead generates ${t}`);
  let a = Wx("input", e.inputNames, n), i = Wx("output", e.outputNames, s), o = a[0].shape[0];
  w.assert(a.length === e.inputs.length, () => `LayersModel has ${e.inputs.length} inputs, but the dataset provides ${a.length} inputs.  (Expected input keys: ${JSON.stringify(e.inputNames)})`), w.assert(i.length === e.outputs.length, () => `LayersModel has ${e.outputs.length} outputs, but the dataset provides ${i.length} outputs.  (Expected output keys: ${JSON.stringify(e.outputNames)})`);
  for (let u = 0; u < a.length; u++)
    w.assert(a[u].shape[0] === o, () => `Batch size mismatch: input ${e.inputNames[u]} has ${a[u].shape[0]}; expected  ${o} based on input ${e.inputNames[0]}.`);
  for (let u = 0; u < i.length; u++)
    w.assert(i[u].shape[0] === o, () => `Batch size mismatch: output ${e.outputNames[u]} has ${i[u].shape[0]}; expected  ${o} based on input ${e.inputNames[0]}.`);
  return { xs: a, ys: i };
}
function Wx(e, t, n) {
  if (n instanceof et)
    return [n];
  if (Array.isArray(n))
    return w.assert(n.length === t.length, () => `Received an array of ${n.length} Tensors, but expected ${t.length} to match the ${e} keys ${t}.`), n;
  {
    let s = [];
    for (let r of t) {
      if (n[r] == null)
        throw new G(`The feature data generated by the dataset lacks the required ${e} key '${r}'.`);
      s.push(n[r]);
    }
    return s;
  }
}
function PB(e) {
  if (e.length === 3)
    throw new Fe("Validation with sample weights is not implemented yet.");
  return { xs: e[0], ys: e[1] };
}
async function zB(e, t, n) {
  let s = n.batchesPerEpoch != null;
  if (w.assert(e.optimizer != null, () => "You must compile a model before training/testing. Use LayersModel.compile(modelCompileConfig)."), w.assert(n != null, () => "For fitDataset(), the 2nd argument (config) is required, but it is not provided in this call."), w.assert(n.epochs != null && n.epochs > 0 && Number.isInteger(n.epochs), () => `For fitDataset(), config.epochs is expected to be a positive integer, but got ${n.epochs}`), w.assert(!s || n.batchesPerEpoch > 0 && Number.isInteger(n.batchesPerEpoch), () => `For fitDataset(), config.batchesPerEpoch is expected to be a positive integer if specified, but got ${n.batchesPerEpoch}`), w.assert(n.validationSplit == null, () => "`validationSplit` is not supported by `fitDataset()`. Use validationData instead."), e.isTraining)
    throw new Error("Cannot start training because another fit() call is ongoing.");
  e.isTraining = true;
  try {
    let r = n.validationData != null, a, i;
    if (r)
      if (Ux(n.validationData))
        w.assert(n.validationBatches == null || n.validationBatches > 0 && Number.isInteger(n.validationBatches), () => `For fitDataset() with dataset-based validation, config.validationBatches is expected not to be provided, or to be a positive integer, but got ${n.validationBatches}`);
      else {
        let g = PB(n.validationData);
        a = g.xs, i = g.ys;
      }
    let o = e.makeTrainFunction(), u = e.getDedupedMetricsNames(), l;
    r ? l = u.slice().concat(u.map((g) => "val_" + g)) : l = u.slice();
    let c = dI(n.callbacks, n.yieldEvery), p = n.verbose == null ? 1 : n.verbose, { callbackList: d, history: h } = pI(c, p, n.epochs, null, null, MB(t, n), null, r, l);
    d.setModel(e), e.history = h, await d.onTrainBegin(), e.stopTraining_ = false;
    let f = n.initialEpoch == null ? 0 : n.initialEpoch, m = await t.iterator();
    for (; f < n.epochs; ) {
      let g = {};
      await d.onEpochBegin(f);
      let b = 0, y = 0;
      for (s || (m = await t.iterator()); !s || b < n.batchesPerEpoch; ) {
        let v = await m.next();
        if (s && v.done) {
          console.warn(`You provided \`batchesPerEpoch\` as ${n.batchesPerEpoch}, but your dataset iterator ran out of data after ${b} batches; interrupting training. Make sure that your dataset can generate at least \`batchesPerEpoch * epochs\` batches (in this case, ${n.batchesPerEpoch * n.epochs} batches). You may need to use the repeat() function when building your dataset.`);
          break;
        }
        if (v.value != null) {
          let { xs: x, ys: k } = kI(e, v.value), I = {};
          I.batch = y, I.size = x[0].shape[0], await d.onBatchBegin(y, I);
          let $ = [];
          if (n.classWeight != null) {
            let P = xI(n.classWeight, e.outputNames);
            for (let A = 0; A < P.length; ++A)
              $.push(await wI(k[A], null, P[A]));
          }
          let R = x.concat(k).concat($), E = o(R);
          De(R);
          for (let P = 0; P < u.length; ++P) {
            let A = u[P], O = E[P];
            I[A] = O, qt(O);
          }
          await d.onBatchEnd(y, I), cI(I), y++, b++;
        }
        if (s ? b >= n.batchesPerEpoch : v.done) {
          if (r) {
            let x;
            Ux(n.validationData) ? x = ht(await e.evaluateDataset(n.validationData, { batches: n.validationBatches })) : x = ht(e.evaluate(a, i, { batchSize: n.validationBatchSize == null ? OB : n.validationBatchSize, verbose: 0 }));
            for (let k = 0; k < e.metricsNames.length; ++k)
              g[`val_${e.metricsNames[k]}`] = x[k];
          }
          break;
        }
        if (e.stopTraining_)
          break;
      }
      if (await d.onEpochEnd(f, g), f++, e.stopTraining_)
        break;
    }
    return await d.onTrainEnd(), await e.history.syncData(), e.history;
  } finally {
    e.isTraining = false;
  }
}
function MB(e, t) {
  let n = null;
  return t.batchesPerEpoch != null ? n = t.batchesPerEpoch : Number.isFinite(e.size) && (n = e.size), n;
}
function Ux(e) {
  return typeof e.iterator == "function";
}
function LB(e) {
  return typeof e.next == "function";
}
async function BB(e, t, n) {
  n = n || {};
  let s = n.batches != null, r = e.testFunction, a = [];
  if (n.verbose > 0)
    throw new Fe("Verbose mode is not implemented yet.");
  w.assert(!s || n.batches > 0 && Number.isInteger(n.batches), () => `Test loop expects \`batches\` to be a positive integer, but received ${JSON.stringify(n.batches)}`);
  let i = LB(t) ? t : await t.iterator(), o = 0, u = 0;
  for (; !s || u < n.batches; ) {
    let l = await i.next();
    if (a = j(() => {
      if (l.value) {
        let { xs: c, ys: p } = kI(e, l.value), d = c.concat(p), h = j(() => r(d));
        if (De(d), u === 0)
          for (let m = 0; m < h.length; ++m)
            a.push(we(0));
        let f = d[0].shape[0];
        for (let m = 0; m < h.length; ++m) {
          let g = h[m], b = a[m];
          a[m] = j(() => ie(a[m], V(f, g))), u > 0 && De(b);
        }
        De(h), o += f, ++u;
      }
      return a;
    }), l.done) {
      s && console.warn(`Your dataset iterator ran out of data during evaluateDataset(). Interrupting evalution. Make sure that your dataset can generate at least \`batches\` batches (in this case, ${n.batches} batches). You may need to use the repeat() function when building your dataset.`);
      break;
    }
  }
  for (let l = 0; l < a.length; ++l) {
    let c = a[l];
    a[l] = xe(a[l], o), De(c);
  }
  return bn(a);
}
function _m(e) {
  w.assert(e > 0 && Number.isInteger(e), () => `batchSize is required to be a positive integer, but got ${e}`);
}
function Ou(e, t, n) {
  return e == null ? [null] : Array.isArray(e) ? e.map((s) => na(s, t, n - t)) : na(e, t, n - t);
}
function Yb(e, t) {
  return j(() => e == null ? null : Array.isArray(e) ? e.map((n) => Yb(n, t)) : rI(e, t.dtype === "int32" ? t : le(t, "int32")));
}
function Am(e, t) {
  let n = [], s = 0, r = null;
  for (; s < e; )
    r = s + t, r >= e && (r = e), n.push([s, r]), s = r;
  return n;
}
async function VB(e, t, n, s, r, a, i, o, u, l, c, p, d, h, f) {
  r == null && (r = 32), a == null && (a = 1), c == null && (c = true), d == null && (d = 0);
  let m = false;
  if (u != null && l != null && (m = true), f != null && (m = true, h == null))
    throw new G("Can only use `validationSteps` when doing step-wise training, i.e., `stepsPerEpoch` must be set.");
  let g = e.checkNumSamples(n, r, h, "steps_per_epoch"), b;
  g != null && (b = ys(0, g)), i == null && (i = 1);
  let { callbackList: y, history: v } = pI(o, i, a, d, g, h, r, m, p);
  y.setModel(e), e.history = v, await y.onTrainBegin(), e.stopTraining_ = false;
  for (let x = d; x < a; ++x) {
    await y.onEpochBegin(x);
    let k = {};
    if (h != null)
      throw new Fe("stepsPerEpoch mode is not implemented yet.");
    {
      if (c === "batch")
        throw new Fe("batch shuffling is not implemneted yet");
      c && w.shuffle(b);
      let I = Zt(b), $ = Am(g, r);
      for (let R = 0; R < $.length; ++R) {
        let E = {};
        if (await y.onBatchBegin(R, E), j(() => {
          let P = $[R][0], A = $[R][1], O = na(I, P, A - P);
          E.batch = R, E.size = A - P;
          let T = Yb(n, O), z = t(T);
          for (let W = 0; W < s.length; ++W) {
            let q = s[W], X = z[W];
            E[q] = X, qt(X);
          }
          if (R === $.length - 1 && m) {
            let W = e.testLoop(u, l, r);
            for (let q = 0; q < s.length; ++q) {
              let X = s[q], Y = W[q];
              qt(Y), k["val_" + X] = Y;
            }
          }
        }), await y.onBatchEnd(R, E), cI(E), e.stopTraining_)
          break;
      }
      I.dispose();
    }
    if (await y.onEpochEnd(x, k), e.stopTraining_)
      break;
  }
  return await y.onTrainEnd(), await e.history.syncData(), e.history;
}
async function WB(e, t, n, s = {}) {
  if (e.isTraining)
    throw new Error("Cannot start training because another fit() call is ongoing.");
  e.isTraining = true;
  let r, a, i, o, u, l, c, p, d;
  try {
    let h = s.batchSize == null ? 32 : s.batchSize;
    _m(h);
    let f = false, m = await e.standardizeUserData(t, n, s.sampleWeight, s.classWeight, f, h);
    r = m[0], a = m[1], d = m[2];
    let g = false, b;
    if (s.validationData != null && s.validationData.length > 0) {
      if (g = true, s.validationData.length === 2)
        u = s.validationData[0], l = s.validationData[1];
      else
        throw s.validationData.length === 3 ? new Fe("validationData including sample weights is not supported yet.") : new G(`When passing validation data, it must contain 2 (valX, valY) or 3 (valX, valY, valSampleWeight) items; ${s.validationData} is invalid.`);
      let E = true, P = await e.standardizeUserData(u, l, null, null, E, h);
      c = P[0], p = P[1], b = c.concat(p);
    } else if (s.validationSplit != null && s.validationSplit > 0 && s.validationSplit < 1) {
      g = true;
      let E = Math.floor(r[0].shape[0] * (1 - s.validationSplit)), P = r[0].shape[0];
      c = Ou(r, E, P), i = r, r = Ou(r, 0, E), p = Ou(a, E, P), o = a, a = Ou(a, 0, E), b = c.concat(p);
    } else
      s.validationSteps != null && (g = true);
    let y = r.concat(a).concat(d);
    e.checkTrainableWeightsConsistency();
    let v = e.makeTrainFunction(), x = e.getDedupedMetricsNames(), k, I;
    g ? (e.makeTestFunction(), k = e.testFunction, I = x.slice().concat(x.map((E) => "val_" + E))) : (k = null, b = [], I = x.slice());
    let $ = dI(s.callbacks, s.yieldEvery);
    return await VB(e, v, y, x, h, s.epochs, s.verbose, $, k, b, s.shuffle, I, s.initialEpoch, null, null);
  } finally {
    e.isTraining = false, ps(r, t), ps(a, n), ps(i, t), ps(o, n), ps(c, u), ps(p, l), d != null && De(d);
  }
}
function SI(e) {
  let t = [];
  e instanceof et && (e = [e]);
  for (let n = 0; n < e.length; ++n) {
    let s = e[n];
    if (s.rank === 1)
      t.push(Ul(s, 1));
    else {
      if (s.rank === 0)
        throw new Error("Expected tensor to be at least 1D, but received a 0D tensor (scalar).");
      t.push(s);
    }
  }
  return t;
}
function ps(e, t) {
  if (e == null)
    return;
  let n = [];
  if (t instanceof et)
    n.push(t.id);
  else if (Array.isArray(t))
    t.forEach((r) => n.push(r.id));
  else if (t != null)
    for (let r in t) {
      let a = t[r];
      n.push(a.id);
    }
  let s = [];
  if (e instanceof et)
    n.indexOf(e.id) === -1 && s.push(e);
  else if (Array.isArray(e))
    e.forEach((r) => {
      n.indexOf(r.id) === -1 && s.push(r);
    });
  else if (e != null)
    for (let r in e) {
      let a = e[r];
      n.indexOf(a.id) === -1 && s.push(a);
    }
  s.forEach((r) => {
    r.isDisposed || r.dispose();
  });
}
function UB(e) {
  return e instanceof et;
}
function Em(e) {
  return Array.isArray(e);
}
function Gx(e) {
  return !UB(e) && !Em(e);
}
function Hx(e, t, n, s = true, r = "") {
  if (t == null || t.length === 0) {
    if (e != null) {
      let i = false;
      if (Em(e) && e.length > 0)
        i = true;
      else if (Gx(e)) {
        for (let o in e)
          if (e.hasOwnProperty(o)) {
            i = true;
            break;
          }
      } else
        i = true;
      if (i)
        throw new G(`Error when checking model ${r} expected no data, but got ${e}`);
    }
    return [];
  }
  if (e == null)
    return t.map((i) => null);
  let a;
  if (Gx(e)) {
    e = e, a = [];
    for (let i of t) {
      if (e[i] == null)
        throw new G(`No data provided for "${i}". Need data for each key in: ${t}`);
      a.push(e[i]);
    }
  } else if (Em(e)) {
    if (e = e, e.length !== t.length)
      throw new G(`Error when checking model ${r}: the Array of Tensors that you are passing to your model is not the size the model expected. Expected to see ${t.length} Tensor(s), but instead got the following list of Tensor(s): ${e}`);
    a = e;
  } else {
    if (e = e, t.length > 1)
      throw new G(`The model ${r} expects ${t.length} Tensor(s), but only received one Tensor. Found: Tensor with shape ${e.shape}`);
    a = [e];
  }
  if (a = SI(a), n != null)
    for (let i = 0; i < t.length; ++i) {
      if (n[i] == null)
        continue;
      let o = a[i];
      if (o.shape.length !== n[i].length)
        throw new G(`Error when checking ${r}: expected ${t[i]} to have ${n[i].length} dimension(s). but got array with shape ${o.shape}`);
      for (let u = 0; u < n[i].length; ++u) {
        if (u === 0 && !s)
          continue;
        let l = o.shape[u], c = n[i][u];
        if (c != null && c >= 0 && l !== c)
          throw new G(`${r} expected a batch of elements where each example has shape [${n[i].slice(1, n[i].length)}] (i.e.,tensor shape [*,${n[i].slice(1, n[i].length)}]) but the ${r} received an input with ${o.shape[0]} examples, each with shape [${o.shape.slice(1, o.shape.length)}] (tensor shape [${o.shape}])`);
      }
    }
  return a;
}
function GB(e, t, n) {
  let s = lr(e.map((a) => a.shape[0]));
  s.sort();
  let r = lr(t.map((a) => a.shape[0]));
  if (r.sort(), s.length > 1)
    throw new G(`All input Tensors (x) should have the same number of samples. Got array shapes: ${JSON.stringify(e.map((a) => a.shape))}`);
  if (r.length > 1)
    throw new G(`All target Tensors (y) should have the same number of samples. Got array shapes: ${JSON.stringify(t.map((a) => a.shape))}`);
  if (s.length > 0 && r.length > 0 && !w.arraysEqual(s, r))
    throw new G(`Input Tensors should have the same number of samples as target Tensors. Found ${s[0]} input sample(s) and ${r[0]} target sample(s).`);
}
function HB(e, t, n) {
  let s = [vi, Up, tl];
  for (let r = 0; r < e.length; ++r) {
    let a = e[r], i = t[r], o = n[r];
    if (i != null) {
      if (i === tl && a.shape[a.shape.length - 1] === 1)
        throw new G(`You are passing a target array of shape ${a.shape} while using a loss 'categorical_crossentropy'. 'categorical_crossentropy'expects targets to be binary matrices (1s and 0s) of shape [samples, classes].`);
      if (s.indexOf(i) !== -1) {
        let u = a.shape.slice(1), l = o.slice(1);
        for (let c = 0; c < u.length; ++c) {
          let p = u[c], d = l[c];
          if (d != null && p !== d)
            throw new G(`A target Tensor with shape ${a.shape} was passed for an output of shape ${o}, while using a loss function that expects targets to have the same shape as the output.`);
        }
      }
    }
  }
}
function qx(e, t, n, s = true, r = "") {
  let a;
  if (Array.isArray(e)) {
    if (e.length !== t.length)
      throw new G(`Error when checking model ${r}: the Array of Tensors that you are passing to your model is not the size the the model expected. Expected to see ${t.length} Tensor(s), but instead got ${e.length} Tensors(s).`);
    a = e;
  } else {
    if (t.length > 1)
      throw new G(`The model expects ${t.length} ${r} Tensors, but only received one Tensor. Found: array with shape ${JSON.stringify(e.shape)}.`);
    a = [e];
  }
  if (n != null)
    for (let i = 0; i < t.length; ++i) {
      if (n[i] == null)
        continue;
      let o = a[i];
      if (o.shape.length !== n[i].length)
        throw new G(`Error when checking ${r}: expected ${t[i]} to have ${n[i].length} dimension(s), but got array with shape ${JSON.stringify(o.shape)}`);
      for (let u = 0; u < n[i].length; ++u) {
        if (u === 0 && !s)
          continue;
        let l = o.shape[u], c = n[i][u];
        if (c != null && c !== l)
          throw new G(`Error when checking ${r}: expected ${t[i]} to have shape ${JSON.stringify(n[i])} but got array with shape ${JSON.stringify(o.shape)}.`);
      }
    }
}
function qB(e, t) {
  if (e == null || Array.isArray(e) && e.length === 0)
    return t.map((s) => []);
  let n;
  if (typeof e == "string" || typeof e == "function")
    n = [e];
  else if (Array.isArray(e) || typeof e == "object")
    n = e;
  else
    throw new TypeError(`Type of metrics argument not understood. Expected an string,function, Array, or Object, found: ${e}`);
  if (Array.isArray(n))
    return t.map((s) => n);
  {
    let s = [];
    for (let r of t) {
      let a = n.hasOwnProperty(r) ? n[r] : [];
      Array.isArray(a) || (a = [a]), s.push(a);
    }
    return s;
  }
}
var jB = "layers-model";
var dr = class extends Is {
  constructor(e) {
    super(e), this.isTraining = false;
  }
  summary(e, t, n = console.log) {
    if (!this.built)
      throw new G("This model has never been called, thus its weights have not been created yet. So no summary can be displayed. Build the model first (e.g., by calling it on some test data).");
    $B(this, e, t, n);
  }
  compile(e) {
    if (e.loss == null && (e.loss = []), this.loss = e.loss, typeof e.optimizer == "string")
      this.optimizer_ = TB(e.optimizer), this.isOptimizerOwned = true;
    else {
      if (!(e.optimizer instanceof _r))
        throw new G("User-defined optimizer must be an instance of tf.Optimizer.");
      this.optimizer_ = e.optimizer, this.isOptimizerOwned = false;
    }
    let t = [];
    if (!Array.isArray(e.loss) && typeof e.loss != "string" && typeof e.loss != "function") {
      e.loss = e.loss;
      for (let a in e.loss)
        if (this.outputNames.indexOf(a) === -1)
          throw new G(`Unknown entry in loss dictionary: "${a}". Only expected the following keys: ${this.outputNames}`);
      for (let a of this.outputNames)
        e.loss[a] == null && console.warn(`Output "${a}" is missing from loss dictionary. We assume this was done on purpose, and we will not be expecting data to be passed to ${a} during training`), t.push(Xf(e.loss[a]));
    } else if (Array.isArray(e.loss)) {
      if (e.loss.length !== this.outputs.length)
        throw new G(`When passing an Array as loss, it should have one entry per model output. The model has ${this.outputs.length} output(s), but you passed loss=${e.loss}.`);
      t = e.loss.map((i) => Xf(i));
    } else {
      let a = Xf(e.loss);
      this.outputs.forEach((i) => {
        t.push(a);
      });
    }
    this.lossFunctions = t, this.feedOutputNames = [], this.feedOutputShapes = [], this.feedLossFns = [];
    for (let a = 0; a < this.outputs.length; ++a) {
      let i = this.internalOutputShapes[a], o = this.outputNames[a];
      this.feedOutputNames.push(o), this.feedOutputShapes.push(i), this.feedLossFns.push(this.lossFunctions[a]);
    }
    let n = [];
    this.metrics = e.metrics, this.metricsNames = ["loss"], this.metricsTensors = [], ta("loss", () => {
      for (let a = 0; a < this.outputs.length; ++a) {
        if (n.indexOf(a) !== -1)
          continue;
        let i = this.lossFunctions[a];
        this.outputs.length > 1 && (this.metricsTensors.push([i, a]), this.metricsNames.push(this.outputNames[a] + "_loss"));
      }
    });
    let s = qB(e.metrics, this.outputNames), r = (a, i, o) => {
      this.outputNames.length > 1 && (i = this.outputNames[a] + "_" + i), this.metricsNames.push(i), this.metricsTensors.push([o, a]);
    };
    ta("metric", () => {
      for (let a = 0; a < this.outputs.length; ++a) {
        if (n.indexOf(a) !== -1)
          continue;
        let i = s[a];
        ((u) => {
          let l = "", c, p, d;
          for (let h of u) {
            if (typeof h == "string" && ["accuracy", "acc", "crossentropy", "ce"].indexOf(h) !== -1) {
              let m = this.internalOutputShapes[a];
              m[m.length - 1] === 1 || this.lossFunctions[a] === Up ? ["accuracy", "acc"].indexOf(h) !== -1 ? p = jb : ["crossentropy", "ce"].indexOf(h) !== -1 && (p = mI) : this.lossFunctions[a] === Dd ? ["accuracy", "acc"].indexOf(h) !== -1 ? p = gI : ["crossentropy", "ce"].indexOf(h) !== -1 && (p = bI) : ["accuracy", "acc"].indexOf(h) !== -1 ? p = Kb : ["crossentropy", "ce"].indexOf(h) !== -1 && (p = Xb);
              let g;
              ["accuracy", "acc"].indexOf(h) !== -1 ? g = "acc" : ["crossentropy", "ce"].indexOf(h) !== -1 && (g = "ce"), d = p, c = l + g;
            } else
              d = NB(h), c = l + Xc(h);
            let f;
            ta(c, () => {
              f = d;
            }), r(a, c, f);
          }
        })(i);
      }
    }), this.collectedTrainableWeights = this.trainableWeights;
  }
  checkTrainableWeightsConsistency() {
    this.collectedTrainableWeights != null && this.trainableWeights.length !== this.collectedTrainableWeights.length && console.warn("Discrepancy between trainableweights and collected trainable weights. Did you set `model.trainable` without calling `model.compile()` afterwards?");
  }
  evaluate(e, t, n = {}) {
    let s = n.batchSize == null ? 32 : n.batchSize;
    _m(s);
    let r = true, a = this.standardizeUserDataXY(e, t, r, s);
    try {
      let i = a[0].concat(a[1]);
      this.makeTestFunction();
      let o = this.testFunction, u = this.testLoop(o, i, s, n.verbose, n.steps);
      return bn(u);
    } finally {
      ps(a[0], e), ps(a[1], t);
    }
  }
  async evaluateDataset(e, t) {
    return this.makeTestFunction(), BB(this, e, t);
  }
  checkNumSamples(e, t, n, s = "steps") {
    let r;
    if (n != null) {
      if (r = null, t != null)
        throw new G(`If ${s} is set, batchSize must be null or undefined.Got batchSize = ${t}`);
    } else if (e != null)
      Array.isArray(e) ? r = e[0].shape[0] : r = e.shape[0];
    else
      throw new G(`Either the input data should have a defined shape, or ${s} shoud be specified.`);
    return r;
  }
  execute(e, t) {
    if (Array.isArray(t) && t.length === 0)
      throw new G("`outputs` is an empty Array, which is not allowed.");
    let n = Array.isArray(t), s = n ? t : [t], r = this.retrieveSymbolicTensors(s), a = new Zr();
    if (e instanceof et && (e = [e]), Array.isArray(e)) {
      if (e.length !== this.inputs.length)
        throw new G(`The number of inputs provided (${e.length}) does not match the number of inputs of this model (${this.inputs.length}).`);
      for (let o = 0; o < this.inputs.length; ++o)
        a.add(this.inputs[o], e[o]);
    } else
      for (let o of this.inputs) {
        let u = e[o.name];
        if (u == null)
          throw new G(`No value is provided for the model's input ${o.name}`);
        a.add(o, u);
      }
    let i = Fu(r, a);
    return n ? i : i[0];
  }
  retrieveSymbolicTensors(e) {
    let t = ma(null, e.length), n = e.length;
    for (let s of this.layers) {
      let r = Array.isArray(s.output) ? s.output : [s.output], a = r.map((i) => i.name);
      for (let i = 0; i < e.length; ++i) {
        let o = a.indexOf(e[i]);
        if (o !== -1 && (t[i] = r[o], n--), n === 0)
          break;
      }
      if (n === 0)
        break;
    }
    if (n > 0) {
      let s = [];
      throw t.forEach((r, a) => {
        r == null && s.push(e[a]);
      }), new G(`Cannot find SymbolicTensors for output name(s): ${JSON.stringify(s)}`);
    }
    return t;
  }
  predictLoop(e, t = 32, n = false) {
    return j(() => {
      let s = this.checkNumSamples(e);
      if (n)
        throw new Fe("Verbose predictLoop() is not implemented yet.");
      let r = Am(s, t), a = this.outputs.map((i) => []);
      for (let i = 0; i < r.length; ++i)
        j(() => {
          let u = r[i][0], l = r[i][1], c = Ou(e, u, l), p = [];
          if (Array.isArray(c))
            for (let h = 0; h < c.length; ++h)
              p.push({ key: this.inputs[h], value: c[h] });
          else
            p.push({ key: this.inputs[0], value: c });
          let d = new Zr(p);
          return Fu(this.outputs, d);
        }).forEach((u, l) => a[l].push(u));
      return bn(a.map((i) => Ot(i, 0)));
    });
  }
  predict(e, t = {}) {
    let n = SI(e);
    qx(n, this.inputNames, this.feedInputShapes, false);
    try {
      let s = t.batchSize == null ? 32 : t.batchSize;
      return _m(s), this.predictLoop(n, s);
    } finally {
      ps(n, e);
    }
  }
  predictOnBatch(e) {
    qx(e, this.inputNames, this.feedInputShapes, true);
    let t = (Array.isArray(e) ? e[0] : e).shape[0];
    return this.predictLoop(e, t);
  }
  standardizeUserDataXY(e, t, n = true, s) {
    if (this.optimizer_ == null)
      throw new fs("You must compile a model before training/testing. Use LayersModel.compile(modelCompileArgs).");
    let r = [];
    for (let a = 0; a < this.feedOutputShapes.length; ++a) {
      let i = this.feedOutputShapes[a];
      this.feedLossFns[a] === Dd ? r.push(i.slice(0, i.length - 1).concat([1])) : r.push(i);
    }
    if (e = Hx(e, this.feedInputNames, this.feedInputShapes, false, "input"), t = Hx(t, this.feedOutputNames, r, false, "target"), GB(e, t, null), HB(t, this.feedLossFns, this.feedOutputShapes), this.stateful && s != null && s > 0 && e[0].shape[0] % s !== 0)
      throw new G(`In a stateful network, you should only pass inputs with a number of samples that is divisible by the batch size ${s}. Found: ${e[0].shape[0]} sample(s).`);
    return [e, t];
  }
  async standardizeUserData(e, t, n, s, r = true, a) {
    let [i, o] = this.standardizeUserDataXY(e, t, r, a);
    if (n != null)
      throw new Error("sample weight is not supported yet.");
    let u = null;
    if (s != null) {
      let l = xI(s, this.outputNames);
      u = [];
      for (let c = 0; c < l.length; ++c)
        u.push(await wI(o[c], null, l[c]));
    }
    return [i, o, u];
  }
  testLoop(e, t, n, s = 0, r) {
    return j(() => {
      let a = this.checkNumSamples(t, n, r, "steps"), i = [];
      if (s > 0)
        throw new Fe("Verbose mode is not implemented yet.");
      if (r != null)
        throw new Fe("steps mode in testLoop() is not implemented yet");
      {
        let o = Am(a, n), u = Zt(ys(0, a));
        for (let l = 0; l < o.length; ++l) {
          let c = o[l][0], p = o[l][1], d = na(u, c, p - c), h = Yb(t, d), f = e(h);
          if (l === 0)
            for (let m = 0; m < f.length; ++m)
              i.push(we(0));
          for (let m = 0; m < f.length; ++m) {
            let g = f[m];
            i[m] = ie(i[m], V(p - c, g));
          }
        }
        for (let l = 0; l < i.length; ++l)
          i[l] = xe(i[l], a);
      }
      return i;
    });
  }
  getDedupedMetricsNames() {
    let e = this.metricsNames, t = [];
    for (let n = 0; n < e.length; ++n) {
      let s = e[n], r = s;
      Cx(e, s) > 1 && (r += `_${Cx(e.slice(0, n), s)}`), t.push(r);
    }
    return t;
  }
  makeTrainFunction() {
    return (e) => {
      let t = [], n = e.slice(0, this.inputs.length), s = e.slice(this.inputs.length, this.inputs.length + this.outputs.length), r = e.slice(this.inputs.length + this.outputs.length, this.inputs.length + this.outputs.length * 2), a = [], i = () => {
        let c = [];
        for (let f = 0; f < this.inputs.length; ++f)
          c.push({ key: this.inputs[f], value: n[f] });
        let p = new Zr(c), d = Fu(this.outputs, p, { training: true }), h;
        for (let f = 0; f < this.lossFunctions.length; ++f) {
          let g = this.lossFunctions[f](s[f], d[f]);
          r[f] != null && (g = FB(g, r[f]));
          let b = It(g);
          t.push(b), f === 0 ? h = g : h = ie(h, g);
        }
        for (let f = 0; f < this.metricsTensors.length; ++f) {
          let m;
          if (this.outputs.length > 1 && f < this.outputs.length)
            m = t[f];
          else {
            let g = this.metricsTensors[f][0], b = this.metricsTensors[f][1];
            m = It(g(s[b], d[b]));
          }
          qt(m), a.push(m);
        }
        return h = It(h), this.calculateLosses().forEach((f) => {
          h = ie(h, f);
        }), h;
      }, o = this.collectedTrainableWeights.map((c) => c.read()), u = true;
      return [this.optimizer_.minimize(i, u, o)].concat(a);
    };
  }
  makeTestFunction() {
    this.testFunction = (e) => j(() => {
      let t = [], n, s = e.slice(0, this.inputs.length), r = e.slice(this.inputs.length, this.inputs.length + this.outputs.length), a = [];
      for (let u = 0; u < this.inputs.length; ++u)
        a.push({ key: this.inputs[u], value: s[u] });
      let i = new Zr(a), o = Fu(this.outputs, i);
      for (let u = 0; u < this.lossFunctions.length; ++u) {
        let l = this.lossFunctions[u], c = It(l(r[u], o[u]));
        u === 0 ? n = c : n = ie(n, c), t.push(n);
      }
      for (let u = 0; u < this.metricsTensors.length; ++u) {
        let l = this.metricsTensors[u][0], c = this.metricsTensors[u][1], p = It(l(r[c], o[c]));
        t.push(p);
      }
      return t;
    });
  }
  async fit(e, t, n = {}) {
    return WB(this, e, t, n);
  }
  async fitDataset(e, t) {
    return zB(this, e, t);
  }
  async trainOnBatch(e, t) {
    let n = await this.standardizeUserData(e, t), s = n[0], r = n[1], i = this.makeTrainFunction()(s.concat(r)), o = [];
    for (let u of i) {
      let l = await u.data();
      o.push(l[0]);
    }
    return De(i), ps(n[0], e), ps(n[1], t), bn(o);
  }
  getNamedWeights(e) {
    let t = [], n = e != null && e.trainableOnly, s = n ? this.trainableWeights : this.weights, r = this.getWeights(n);
    for (let a = 0; a < s.length; ++a)
      n && !s[a].trainable || t.push({ name: s[a].originalName, tensor: r[a] });
    return t;
  }
  set stopTraining(e) {
    this.stopTraining_ = e;
  }
  get stopTraining() {
    return this.stopTraining_;
  }
  get optimizer() {
    return this.optimizer_;
  }
  set optimizer(e) {
    this.optimizer_ !== e && (this.optimizer_ = e, this.isOptimizerOwned = false);
  }
  dispose() {
    let e = super.dispose();
    if (e.refCountAfterDispose === 0 && this.optimizer != null && this.isOptimizerOwned) {
      let t = bm().numTensors;
      this.optimizer_.dispose(), e.numDisposedVariables += t - bm().numTensors;
    }
    return e;
  }
  getLossIdentifiers() {
    let e;
    if (typeof this.loss == "string")
      e = Vs(this.loss);
    else if (Array.isArray(this.loss)) {
      for (let t of this.loss)
        if (typeof t != "string")
          throw new Error("Serialization of non-string loss is not supported.");
      e = this.loss.map((t) => Vs(t));
    } else {
      let t = Object.keys(this.loss);
      e = {};
      let n = this.loss;
      for (let s of t)
        if (typeof n[s] == "string")
          e[s] = Vs(n[s]);
        else
          throw new Error("Serialization of non-string loss is not supported.");
    }
    return e;
  }
  getMetricIdentifiers() {
    if (typeof this.metrics == "string" || typeof this.metrics == "function")
      return [Vs(Xc(this.metrics))];
    if (Array.isArray(this.metrics))
      return this.metrics.map((e) => Vs(Xc(e)));
    {
      let e = {};
      for (let t in this.metrics)
        e[t] = Vs(Xc(this.metrics[t]));
      return e;
    }
  }
  getTrainingConfig() {
    return { loss: this.getLossIdentifiers(), metrics: this.getMetricIdentifiers(), optimizer_config: { class_name: this.optimizer.getClassName(), config: this.optimizer.getConfig() } };
  }
  loadTrainingConfig(e) {
    if (e.weighted_metrics != null)
      throw new Error("Loading weight_metrics is not supported yet.");
    if (e.loss_weights != null)
      throw new Error("Loading loss_weights is not supported yet.");
    if (e.sample_weight_mode != null)
      throw new Error("Loading sample_weight_mode is not supported yet.");
    let t = nl(e.optimizer_config), n = gs(t), s;
    if (typeof e.loss == "string")
      s = Xr(e.loss);
    else if (Array.isArray(e.loss))
      s = e.loss.map((a) => Xr(a));
    else if (e.loss != null) {
      s = {};
      for (let a in e.loss)
        s[a] = Xr(e.loss[a]);
    }
    let r;
    if (Array.isArray(e.metrics))
      r = e.metrics.map((a) => Xr(a));
    else if (e.metrics != null) {
      r = {};
      for (let a in e.metrics)
        r[a] = Xr(e.metrics[a]);
    }
    this.compile({ loss: s, metrics: r, optimizer: n });
  }
  async save(e, t) {
    if (typeof e == "string") {
      let u = An.getSaveHandlers(e);
      if (u.length === 0)
        throw new G(`Cannot find any save handlers for URL '${e}'`);
      if (u.length > 1)
        throw new G(`Found more than one (${u.length}) save handlers for URL '${e}'`);
      e = u[0];
    }
    if (e.save == null)
      throw new G("LayersModel.save() cannot proceed because the IOHandler provided does not have the `save` attribute defined.");
    let n = await An.encodeWeights(this.getNamedWeights(t)), s = false, r = null, i = { modelTopology: this.toJSON(r, s), format: jB, generatedBy: `TensorFlow.js tfjs-layers v${vI}`, convertedBy: null };
    if ((t == null ? false : t.includeOptimizer) && this.optimizer != null) {
      i.trainingConfig = this.getTrainingConfig();
      let u = "optimizer", { data: l, specs: c } = await An.encodeWeights(await this.optimizer.getWeights(), u);
      n.specs.push(...c), n.data = An.concatenateArrayBuffers([n.data, l]);
    }
    return this.userDefinedMetadata != null && (Vx(this.userDefinedMetadata, this.name, true), i.userDefinedMetadata = this.userDefinedMetadata), i.weightData = n.data, i.weightSpecs = n.specs, e.save(i);
  }
  setUserDefinedMetadata(e) {
    Vx(e, this.name), this.userDefinedMetadata = e;
  }
  getUserDefinedMetadata() {
    return this.userDefinedMetadata;
  }
};
dr.className = "Model";
re.registerClass(dr);
var II = class extends dr {
};
II.className = "Functional";
re.registerClass(II);
async function KB(e, t) {
  "modelTopology" in e || (e = { modelTopology: e }), e = e;
  let n = e.modelTopology;
  n.model_config != null && (n = n.model_config);
  let s = nl(n), r = gs(s, t);
  if (e.weightsManifest != null) {
    let a = await An.loadWeights(e.weightsManifest, e.pathPrefix, r.weights.map((o) => o.originalName)), i = {};
    for (let o of r.weights)
      i[o.originalName] = a[o.originalName];
    r.loadWeights(i), De(a);
  }
  return r;
}
async function XB(e, t) {
  if (t == null && (t = {}), typeof e == "string") {
    let n = An.getLoadHandlers(e, t);
    if (n.length === 0)
      n.push(An.browserHTTPRequest(e, t));
    else if (n.length > 1)
      throw new G(`Found more than one (${n.length}) load handlers for URL '${e}'`);
    e = n[0];
  }
  return YB(e, void 0, t);
}
async function YB(e, t, n) {
  if (n == null && (n = {}), e.load == null)
    throw new G("Cannot proceed with model loading because the IOHandler provided does not have the `load` method implemented.");
  let s = await e.load(), r = s.modelTopology;
  r.model_config != null && (r = r.model_config);
  let a = n.strict == null ? true : n.strict, i = s.weightData != null && s.weightSpecs != null && a, o = gs(nl(r), t, i), u = s.trainingConfig;
  if (u != null && o.loadTrainingConfig(u), s.userDefinedMetadata != null && o.setUserDefinedMetadata(s.userDefinedMetadata), s.weightData != null) {
    if (s.weightSpecs == null)
      throw new G("LayersModel artifacts contains weight data, but not weight specs. Therefore loading of weights cannot proceed.");
    let { modelWeights: l, optimizerWeights: c } = QB(s.weightData, s.weightSpecs);
    o.loadWeights(l, a), o.optimizer != null && c.length > 0 && await o.optimizer.setWeights(c), De(l), De(c.map((p) => p.tensor));
  }
  return o;
}
function QB(e, t) {
  let n = An.decodeWeights(e, t), s = {}, r = [];
  return t.forEach((a) => {
    a.group === "optimizer" ? r.push({ name: a.name, tensor: n[a.name] }) : s[a.name] = n[a.name];
  }), { modelWeights: s, optimizerWeights: r };
}
var Rm = class extends dr {
  constructor(e) {
    if (super({ inputs: [], outputs: [] }), e = e || {}, this.trainable = true, this.built = false, this.name = e.name != null ? e.name : Ep("sequential_"), e.layers != null)
      for (let t of e.layers)
        this.add(t);
  }
  checkShape(e) {
    if (e.inboundNodes[0].outputTensors[0].shape.some((n) => n < 0))
      throw new G(`Negative dimension size caused by adding layer ${e.name} with input shape [${e.inboundNodes[0].inputTensors[0].shape}]`);
  }
  add(e) {
    let t = e instanceof Rm || e instanceof dr, n;
    if (t) {
      if (n = e, n.outputs.length !== 1)
        throw new G("All layers in a Sequential model should have a single output tensor. For multi-output layers, use the functional API.");
      if (n.inputs.length !== 1)
        throw new G("All layers in a Sequential model should have a single input tensor. For multi-input layers, use the functional API.");
    }
    if (this.outputs.length === 0) {
      if (e.inboundNodes.length === 0) {
        if (e.batchInputShape == null)
          throw new G("The first layer in a Sequential model must get an `inputShape` or `batchInputShape` argument.");
        let s = oI({ batchShape: e.batchInputShape, dtype: e.dtype, name: e.name + "_input" });
        e.apply(s);
      }
      if (t)
        this.outputs = n.outputs, this.inputs = n.inputs;
      else {
        if (e.inboundNodes.length !== 1)
          throw new G(`A layer added to a Sequential model must not already be connected somewhere else. LayersModel received layer ${e.name} which has ${e.inboundNodes.length} pre-existing inbound connections.`);
        if (e.inboundNodes[0].outputTensors.length !== 1)
          throw new G("All layers in a Sequential model should have a single output tensor. For multi-output layers, use the functional API.");
        this.checkShape(e), this.outputs = [e.inboundNodes[0].outputTensors[0]], this.inputs = iI(this.outputs[0]);
      }
      this.inboundNodes = [], new Vp({ outboundLayer: this, inboundLayers: [], nodeIndices: [], tensorIndices: [], inputTensors: this.inputs, outputTensors: this.outputs, inputMasks: ma(null, this.inputs.length), outputMasks: [null], inputShapes: this.inputs.map((s) => s.shape), outputShapes: this.outputs[0].shape });
    } else {
      let s = e.apply(this.outputs[0]);
      if (Array.isArray(s))
        throw new TypeError("All layers in a Sequential model should have a single output tensor. For multi-output layers, use the functional API.");
      this.checkShape(e), this.outputs = [s], this.inboundNodes[0].outputTensors = this.outputs, this.inboundNodes[0].outputShapes = [this.outputs[0].shape];
    }
    this.layers.push(e), this.built = false;
  }
  pop() {
    if (this.layers.length === 0)
      throw new TypeError("There are no layers in the model.");
    if (this.layers.pop(), this.layers.length === 0)
      this.outputs = [], this.inboundNodes = [], this.outboundNodes = [];
    else {
      let e = this.layers.length - 1;
      this.layers[e].outboundNodes = [], this.outputs = [this.layers[e].output], this.inboundNodes[0].outputTensors = this.outputs, this.inboundNodes[0].outputShapes = [this.outputs[0].shape];
    }
  }
  call(e, t) {
    return this.model == null && this.build(), this.model.call(e, t);
  }
  build(e) {
    if (nt(e), this.inputs.length === 0 || this.outputs.length === 0)
      throw new TypeError("Sequential model cannot be built: model is empty. Add some layers first.");
    this.model = new dr({ inputs: this.inputs, outputs: this.outputs[0], name: this.name + "_model" }), this.model.trainable = this.trainable, this.supportsMasking = this.model.supportsMasking, this.inputLayers = this.model.inputLayers, this.inputLayersNodeIndices = this.model.inputLayersNodeIndices, this.inputLayersTensorIndices = this.model.inputLayersTensorIndices, this.outputLayers = this.model.outputLayers, this.outputLayersNodeIndices = this.model.outputLayersNodeIndices, this.outputLayersTensorIndices = this.model.outputLayersTensorIndices, this.nodesByDepth = this.model.nodesByDepth, this.containerNodes = this.model.containerNodes, this.outputNames = this.model.outputNames, this.inputNames = this.model.inputNames, this.built = true;
  }
  countParams() {
    return this.built || this.build(), super.countParams();
  }
  summary(e, t, n = console.log) {
    this.built || this.build(), super.summary(e, t, n);
  }
  setWeights(e) {
    this.model == null && this.build(), this.model.setWeights(e);
  }
  evaluate(e, t, n = {}) {
    if (!this.built)
      throw new fs("The model needs to be compiled before being used.");
    return this.model.evaluate(e, t, n);
  }
  async evaluateDataset(e, t) {
    if (!this.built)
      throw new fs("The model needs to be compiled before being used.");
    return this.model.evaluateDataset(e, t);
  }
  predict(e, t = {}) {
    return this.model == null && this.build(), this.model.predict(e, t);
  }
  predictOnBatch(e) {
    return this.model == null && this.build(), this.model.predictOnBatch(e);
  }
  compile(e) {
    this.build(), this.model.compile(e), this.optimizer_ = this.model.optimizer, this.isOptimizerOwned = this.model.isOptimizerOwned, this.loss = this.model.loss, this.metrics = this.model.metrics, this.metricsTensors = this.model.metricsTensors, this.metricsNames = this.model.metricsNames;
  }
  get optimizer() {
    return this.model == null ? void 0 : this.model.optimizer;
  }
  set optimizer(e) {
    this.model.optimizer = e;
  }
  async fit(e, t, n = {}) {
    if (!this.built)
      throw new fs("The model needs to be compiled before being used.");
    return this.model.fit(e, t, n);
  }
  async fitDataset(e, t) {
    if (!this.built)
      throw new fs("The model needs to be compiled before being used.");
    return this.model.fitDataset(e, t);
  }
  async trainOnBatch(e, t) {
    return this.model.trainOnBatch(e, t);
  }
  static fromConfig(e, t, n = {}, s = false) {
    let r, a = {};
    if (t instanceof Array) {
      if (t[0].className == null || t[0].className === "Merge")
        throw new G("Legacy serialization format not supported yet.");
      r = t;
    } else
      w.assert(t.layers != null, () => "When the config data for a Sequential model is not an Array, it must be an Object that contains the 'layers' field."), r = t.layers, delete t.layers, a = t;
    let i = new e(a);
    if (!(i instanceof Rm))
      throw new Fe(`Sequential.fromConfig called on non-Sequential input: ${i}`);
    for (let o of r) {
      let l = gs(o, void 0, s);
      s && l.setFastWeightInitDuringBuild(true), i.add(l);
    }
    return i;
  }
  set stopTraining(e) {
    if (this.model == null)
      throw new G("Cannot set the stopTraining property of a sequential model before it is compiled.");
    this.model.stopTraining = e;
  }
  get stopTraining() {
    if (this.model == null)
      throw new G("Cannot get the stopTraining property of a sequential model before it is compiled.");
    return this.model.stopTraining;
  }
  getConfig() {
    let e = [];
    for (let t of this.layers) {
      let n = {};
      n.className = t.getClassName(), n.config = t.getConfig(), e.push(n);
    }
    return { name: this.name, layers: e };
  }
};
var Qb = Rm;
Qb.className = "Sequential";
re.registerClass(Qb);
function rhe(e) {
  return new dr(e);
}
function ahe(e) {
  return new Qb(e);
}
function ihe(e, t) {
  return t == null && (t = {}), XB(e, t);
}
function ZB(e) {
  return oI(e);
}
function ohe(e, t) {
  Hb.registerCallbackConstructor(e, t);
}
var kn = class extends re.Serializable {
  getConfig() {
    return {};
  }
};
var CI = class extends kn {
  apply(e, t = 1) {
    return dz(e, t);
  }
};
CI.className = "elu";
re.registerClass(CI);
var NI = class extends kn {
  apply(e) {
    return $S(e);
  }
};
NI.className = "selu";
re.registerClass(NI);
var TI = class extends kn {
  apply(e) {
    return Xs(e);
  }
};
TI.className = "relu";
re.registerClass(TI);
var $I = class extends kn {
  apply(e) {
    return j(() => Ip(6, Xs(e)));
  }
};
$I.className = "relu6";
re.registerClass($I);
var _I = class extends kn {
  apply(e) {
    return e;
  }
};
_I.className = "linear";
re.registerClass(_I);
var AI = class extends kn {
  apply(e) {
    return Hs(e);
  }
};
AI.className = "sigmoid";
re.registerClass(AI);
var EI = class extends kn {
  apply(e) {
    return hz(e);
  }
};
EI.className = "hardSigmoid";
re.registerClass(EI);
var RI = class extends kn {
  apply(e) {
    return Bl(e);
  }
};
RI.className = "softplus";
re.registerClass(RI);
var DI = class extends kn {
  apply(e) {
    return pz(e);
  }
};
DI.className = "softsign";
re.registerClass(DI);
var FI = class extends kn {
  apply(e) {
    return Yu(e);
  }
};
FI.className = "tanh";
re.registerClass(FI);
var Zb = class extends kn {
  apply(e, t = -1) {
    return gb(e, t);
  }
};
Zb.className = "softmax";
re.registerClass(Zb);
var OI = class extends kn {
  apply(e, t = -1) {
    return xS(e, t);
  }
};
OI.className = "logSoftmax";
re.registerClass(OI);
var PI = class extends kn {
  apply(e, t = 1) {
    return j(() => V(Hs(V(e, t)), e));
  }
};
PI.className = "swish";
re.registerClass(PI);
var zI = class extends kn {
  apply(e) {
    return j(() => V(e, Yu(Bl(e))));
  }
};
zI.className = "mish";
re.registerClass(zI);
function br(e) {
  return e.getClassName();
}
function Yf(e, t = {}) {
  return Wl(e, re.SerializationMap.getMap().classNameMap, t, "activation");
}
function yr(e) {
  if (e == null) {
    let t = {};
    return t.className = "linear", t.config = {}, Yf(t);
  }
  if (typeof e == "string") {
    let t = {};
    return t.className = e, t.config = {}, Yf(t);
  } else
    return e instanceof kn ? e : Yf(e);
}
function Jb(e) {
  if (e != null && typeof e != "object")
    throw new Error(`Argument to L1L2 regularizer's constructor is expected to be an object, but received: ${e}`);
}
var MI = class extends re.Serializable {
};
var jl = class extends MI {
  constructor(e) {
    super(), Jb(e), this.l1 = e == null || e.l1 == null ? 0.01 : e.l1, this.l2 = e == null || e.l2 == null ? 0.01 : e.l2, this.hasL1 = this.l1 !== 0, this.hasL2 = this.l2 !== 0;
  }
  apply(e) {
    return j(() => {
      let t = $t([1]);
      return this.hasL1 && (t = ie(t, ve(V(this.l1, Lt(e))))), this.hasL2 && (t = ie(t, ve(V(this.l2, Gl(e))))), U(t, []);
    });
  }
  getConfig() {
    return { l1: this.l1, l2: this.l2 };
  }
  static fromConfig(e, t) {
    return new e({ l1: t.l1, l2: t.l2 });
  }
};
jl.className = "L1L2";
re.registerClass(jl);
function JB(e) {
  return Jb(e), new jl({ l1: e != null ? e.l1 : null, l2: 0 });
}
function eV(e) {
  return Jb(e), new jl({ l2: e != null ? e.l2 : null, l1: 0 });
}
var jx = { l1l2: "L1L2" };
function it(e) {
  return _b(e);
}
function Kx(e, t = {}) {
  return Wl(e, re.SerializationMap.getMap().classNameMap, t, "regularizer");
}
function mt(e) {
  if (e == null)
    return null;
  if (typeof e == "string") {
    let n = { className: e in jx ? jx[e] : e, config: {} };
    return Kx(n);
  } else
    return e instanceof MI ? e : Kx(e);
}
var ey = class extends He {
  constructor(e) {
    super(e == null ? {} : e), this.supportsMasking = true, e != null && (this.maxValue = e.maxValue);
  }
  call(e, t) {
    e = Oe(e);
    let n = Xs(e);
    return this.maxValue != null && (n = Vn(n, 0, this.maxValue)), n;
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = { maxValue: this.maxValue }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
ey.className = "ReLU";
re.registerClass(ey);
var ty = class extends He {
  constructor(e) {
    super(e == null ? {} : e), this.DEFAULT_ALPHA = 0.3, e == null && (e = {}), this.alpha = e.alpha == null ? this.DEFAULT_ALPHA : e.alpha;
  }
  call(e, t) {
    let n = Oe(e);
    return ab(n, this.alpha);
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = { alpha: this.alpha }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
ty.className = "LeakyReLU";
re.registerClass(ty);
var ny = class extends He {
  constructor(e) {
    if (super(e == null ? {} : e), this.DEFAULT_ALPHA_INITIALIZER = "zeros", e == null && (e = {}), this.supportsMasking = true, this.alphaInitializer = ft(e.alphaInitializer || this.DEFAULT_ALPHA_INITIALIZER), this.alphaRegularizer = mt(e.alphaRegularizer), this.alphaConstraint = zt(e.alphaConstraint), e.sharedAxes == null)
      this.sharedAxes = null;
    else if (Array.isArray(e.sharedAxes))
      this.sharedAxes = e.sharedAxes;
    else if (typeof e.sharedAxes == "number")
      this.sharedAxes = [e.sharedAxes];
    else
      throw new G(`Expected sharedAxes to be a number or an array of numbers, but got ${e.sharedAxes}`);
  }
  build(e) {
    e = nt(e);
    let t = e.slice(1);
    if (this.sharedAxes != null)
      for (let s of this.sharedAxes)
        t[s - 1] = 1;
    this.alpha = this.addWeight("alpha", t, "float32", this.alphaInitializer, this.alphaRegularizer, true, this.alphaConstraint);
    let n = {};
    if (this.sharedAxes != null)
      for (let s = 1; s < e.length; ++s)
        n[s] = e[s];
    this.inputSpec = [new Ft({ ndim: e.length, axes: n })], this.built = true;
  }
  call(e, t) {
    return e = Oe(e), db(e, this.alpha.read());
  }
  getConfig() {
    let e = { alphaInitializer: yt(this.alphaInitializer), alphaRegularizer: it(this.alphaRegularizer), alphaConstraint: Pt(this.alphaConstraint), sharedAxes: this.sharedAxes }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
ny.className = "PReLU";
re.registerClass(ny);
var sy = class extends He {
  constructor(e) {
    if (super(e == null ? {} : e), this.DEFAULT_ALPHA = 1, e == null && (e = {}), e.alpha != null && e.alpha !== this.DEFAULT_ALPHA)
      throw new Fe(`Non-default alpha value (${e.alpha}) is not supported by the ELU layer yet.`);
    this.alpha = e.alpha == null ? this.DEFAULT_ALPHA : e.alpha;
  }
  call(e, t) {
    let n = Oe(e);
    return wp(n);
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = { alpha: this.alpha }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
sy.className = "ELU";
re.registerClass(sy);
var ry = class extends He {
  constructor(e) {
    super(e == null ? {} : e), this.DEFAULT_THETA = 1, e == null && (e = {}), this.theta = e.theta == null ? this.DEFAULT_THETA : e.theta;
  }
  call(e, t) {
    let n = Oe(e);
    return V(n, le(Un(n, this.theta), "float32"));
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = { theta: this.theta }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
ry.className = "ThresholdedReLU";
re.registerClass(ry);
var ay = class extends He {
  constructor(e) {
    super(e == null ? {} : e), this.DEFAULT_AXIS = 1, e == null && (e = {}), this.softmax = new Zb().apply, this.axis = e.axis == null ? this.DEFAULT_AXIS : e.axis;
  }
  call(e, t) {
    let n = Oe(e);
    return this.softmax(n, this.axis);
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = { axis: this.axis }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
ay.className = "Softmax";
re.registerClass(ay);
function Ji(e, t, n) {
  if (typeof e == "number")
    return ma(e, t);
  if (e.length !== t)
    throw new G(`The ${n} argument must be an integer or tuple of ${t} integers. Received: ${e.length} elements.`);
  for (let s = 0; s < t; ++s) {
    let r = e[s];
    if (!oz(r))
      throw new G(`The ${n} argument must be an integer or tuple of ${t} integers. Received: ${JSON.stringify(e)} including a non-integer number ${r}`);
  }
  return e;
}
function bs(e, t, n, s, r = 1) {
  if (e == null)
    return e;
  let a = t + (t - 1) * (r - 1), i;
  return n === "same" ? i = e : i = e - a + 1, Math.floor((i + s - 1) / s);
}
function Ns(e, t, n, s) {
  if (e == null)
    return null;
  if (s === "valid")
    e = e * t + gr([n - t, 0]);
  else if (s === "same")
    e = e * t;
  else
    throw new G(`Unsupport padding mode: ${s}.`);
  return e;
}
function iy(e, t) {
  return j(() => (Ct(t), t === "channelsFirst" ? Ge(e, [0, 2, 3, 1]) : e));
}
function LI(e, t) {
  return j(() => (Ct(t), t === "channelsFirst" ? Ge(e, [0, 2, 3, 4, 1]) : e));
}
function tV(e, t, n, s = 1, r = "valid", a, i = 1) {
  return j(() => {
    if (a == null && (a = vs()), Ct(a), e.shape.length !== 3)
      throw new G(`The input of a conv1dWithBias operation should be 3, but is ${e.shape.length} instead.`);
    if (t.shape.length !== 3)
      throw new G(`The kernel for a conv1dWithBias operation should be 3, but is ${t.shape.length} instead`);
    if (n != null && n.shape.length !== 1)
      throw new G(`The bias for a conv1dWithBias operation should be 1, but is ${t.shape.length} instead`);
    if (a === "channelsFirst" && (e = Ge(e, [0, 2, 1])), r === "causal")
      throw new Fe("The support for CAUSAL padding mode in conv1dWithBias is not implemented yet.");
    let o = uS(e, t, s, r === "same" ? "same" : "valid", "NWC", i);
    return n != null && (o = ks(o, n)), o;
  });
}
function Xx(e, t, n, s = [1, 1], r = "valid", a, i, o = null) {
  return j(() => {
    if (a == null && (a = vs()), Ct(a), e.rank !== 3 && e.rank !== 4)
      throw new G(`conv2dWithBiasActivation expects input to be of rank 3 or 4, but received ${e.rank}.`);
    if (t.rank !== 3 && t.rank !== 4)
      throw new G(`conv2dWithBiasActivation expects kernel to be of rank 3 or 4, but received ${e.rank}.`);
    let u = iy(e, a);
    if (r === "causal")
      throw new Fe("The support for CAUSAL padding mode in conv1dWithBias is not implemented yet.");
    return u = fa.conv2d({ x: u, filter: t, strides: s, pad: r === "same" ? "same" : "valid", dilations: i, dataFormat: "NHWC", bias: n, activation: o }), a === "channelsFirst" && (u = Ge(u, [0, 3, 1, 2])), u;
  });
}
function nV(e, t, n, s = [1, 1, 1], r = "valid", a, i) {
  return j(() => {
    if (a == null && (a = vs()), Ct(a), e.rank !== 4 && e.rank !== 5)
      throw new G(`conv3dWithBias expects input to be of rank 4 or 5, but received ${e.rank}.`);
    if (t.rank !== 4 && t.rank !== 5)
      throw new G(`conv3dWithBias expects kernel to be of rank 4 or 5, but received ${e.rank}.`);
    let o = LI(e, a);
    if (r === "causal")
      throw new Fe("The support for CAUSAL padding mode in conv3dWithBias is not implemented yet.");
    return o = cS(o, t, s, r === "same" ? "same" : "valid", "NDHWC", i), n != null && (o = ks(o, n)), a === "channelsFirst" && (o = Ge(o, [0, 4, 1, 2, 3])), o;
  });
}
var oy = class extends He {
  constructor(e, t) {
    if (super(t), this.bias = null, this.DEFAULT_KERNEL_INITIALIZER = "glorotNormal", this.DEFAULT_BIAS_INITIALIZER = "zeros", oy.verifyArgs(t), this.rank = e, Vt(this.rank, "rank"), this.rank !== 1 && this.rank !== 2 && this.rank !== 3)
      throw new Fe(`Convolution layer for rank other than 1, 2, or 3 (${this.rank}) is not implemented yet.`);
    if (this.kernelSize = Ji(t.kernelSize, e, "kernelSize"), this.strides = Ji(t.strides == null ? 1 : t.strides, e, "strides"), this.padding = t.padding == null ? "valid" : t.padding, Gn(this.padding), this.dataFormat = t.dataFormat == null ? "channelsLast" : t.dataFormat, Ct(this.dataFormat), this.activation = yr(t.activation), this.useBias = t.useBias == null ? true : t.useBias, this.biasInitializer = ft(t.biasInitializer || this.DEFAULT_BIAS_INITIALIZER), this.biasConstraint = zt(t.biasConstraint), this.biasRegularizer = mt(t.biasRegularizer), this.activityRegularizer = mt(t.activityRegularizer), this.dilationRate = Ji(t.dilationRate == null ? 1 : t.dilationRate, e, "dilationRate"), this.rank === 1 && Array.isArray(this.dilationRate) && this.dilationRate.length !== 1)
      throw new G(`dilationRate must be a number or an array of a single number for 1D convolution, but received ${JSON.stringify(this.dilationRate)}`);
    if (this.rank === 2) {
      if (typeof this.dilationRate == "number")
        this.dilationRate = [this.dilationRate, this.dilationRate];
      else if (this.dilationRate.length !== 2)
        throw new G(`dilationRate must be a number or array of two numbers for 2D convolution, but received ${JSON.stringify(this.dilationRate)}`);
    } else if (this.rank === 3) {
      if (typeof this.dilationRate == "number")
        this.dilationRate = [this.dilationRate, this.dilationRate, this.dilationRate];
      else if (this.dilationRate.length !== 3)
        throw new G(`dilationRate must be a number or array of three numbers for 3D convolution, but received ${JSON.stringify(this.dilationRate)}`);
    }
  }
  static verifyArgs(e) {
    if (Cs("kernelSize" in e, "required key 'kernelSize' not in config"), typeof e.kernelSize != "number" && !Ab(e.kernelSize, "number", 1, 3))
      throw new G(`BaseConv expects config.kernelSize to be number or number[] with length 1, 2, or 3, but received ${JSON.stringify(e.kernelSize)}.`);
  }
  getConfig() {
    let e = { kernelSize: this.kernelSize, strides: this.strides, padding: this.padding, dataFormat: this.dataFormat, dilationRate: this.dilationRate, activation: br(this.activation), useBias: this.useBias, biasInitializer: yt(this.biasInitializer), biasRegularizer: it(this.biasRegularizer), activityRegularizer: it(this.activityRegularizer), biasConstraint: Pt(this.biasConstraint) }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
var Kl = class extends oy {
  constructor(e, t) {
    super(e, t), this.kernel = null, Kl.verifyArgs(t), this.filters = t.filters, Vt(this.filters, "filters"), this.kernelInitializer = ft(t.kernelInitializer || this.DEFAULT_KERNEL_INITIALIZER), this.kernelConstraint = zt(t.kernelConstraint), this.kernelRegularizer = mt(t.kernelRegularizer);
  }
  build(e) {
    e = nt(e);
    let t = this.dataFormat === "channelsFirst" ? 1 : e.length - 1;
    if (e[t] == null)
      throw new G(`The channel dimension of the input should be defined. Found ${e[t]}`);
    let n = e[t], s = this.kernelSize.concat([n, this.filters]);
    this.kernel = this.addWeight("kernel", s, null, this.kernelInitializer, this.kernelRegularizer, true, this.kernelConstraint), this.useBias && (this.bias = this.addWeight("bias", [this.filters], null, this.biasInitializer, this.biasRegularizer, true, this.biasConstraint)), this.inputSpec = [{ ndim: this.rank + 2, axes: { [t]: n } }], this.built = true;
  }
  call(e, t) {
    return j(() => {
      e = Oe(e);
      let n, s = this.bias == null ? null : this.bias.read(), r = ZS(this.activation.getClassName());
      if (r != null && this.rank === 2)
        n = Xx(e, this.kernel.read(), s, this.strides, this.padding, this.dataFormat, this.dilationRate, r);
      else {
        if (this.rank === 1)
          n = tV(e, this.kernel.read(), s, this.strides[0], this.padding, this.dataFormat, this.dilationRate[0]);
        else if (this.rank === 2)
          n = Xx(e, this.kernel.read(), s, this.strides, this.padding, this.dataFormat, this.dilationRate);
        else if (this.rank === 3)
          n = nV(e, this.kernel.read(), s, this.strides, this.padding, this.dataFormat, this.dilationRate);
        else
          throw new Fe("convolutions greater than 3D are not implemented yet.");
        this.activation != null && (n = this.activation.apply(n));
      }
      return n;
    });
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = [], n = this.dataFormat === "channelsLast" ? e.slice(1, e.length - 1) : e.slice(2);
    for (let r = 0; r < n.length; ++r) {
      let a = bs(n[r], this.kernelSize[r], this.padding, this.strides[r], typeof this.dilationRate == "number" ? this.dilationRate : this.dilationRate[r]);
      t.push(a);
    }
    let s = [e[0]];
    return this.dataFormat === "channelsLast" ? (s = s.concat(t), s.push(this.filters)) : (s.push(this.filters), s = s.concat(t)), s;
  }
  getConfig() {
    let e = { filters: this.filters, kernelInitializer: yt(this.kernelInitializer), kernelRegularizer: it(this.kernelRegularizer), kernelConstraint: Pt(this.kernelConstraint) }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
  static verifyArgs(e) {
    if (!("filters" in e) || typeof e.filters != "number" || e.filters < 1)
      throw new G(`Convolution layer expected config.filters to be a 'number' > 0 but got ${JSON.stringify(e.filters)}`);
  }
};
var BI = class extends Kl {
  constructor(e) {
    super(2, e), BI.verifyArgs(e);
  }
  getConfig() {
    let e = super.getConfig();
    return delete e.rank, e;
  }
  static verifyArgs(e) {
    if (typeof e.kernelSize != "number" && !Ab(e.kernelSize, "number", 1, 2))
      throw new G(`Conv2D expects config.kernelSize to be number or number[] with length 1 or 2, but received ${JSON.stringify(e.kernelSize)}.`);
  }
};
var Gp = BI;
Gp.className = "Conv2D";
re.registerClass(Gp);
var VI = class extends Kl {
  constructor(e) {
    super(3, e), VI.verifyArgs(e);
  }
  getConfig() {
    let e = super.getConfig();
    return delete e.rank, e;
  }
  static verifyArgs(e) {
    if (typeof e.kernelSize != "number" && !(Array.isArray(e.kernelSize) && (e.kernelSize.length === 1 || e.kernelSize.length === 3)))
      throw new G(`Conv3D expects config.kernelSize to be number or [number, number, number], but received ${JSON.stringify(e.kernelSize)}.`);
  }
};
var Hp = VI;
Hp.className = "Conv3D";
re.registerClass(Hp);
var uy = class extends Gp {
  constructor(e) {
    if (super(e), this.inputSpec = [new Ft({ ndim: 4 })], this.padding !== "same" && this.padding !== "valid")
      throw new G(`Conv2DTranspose currently supports only padding modes 'same' and 'valid', but received padding mode ${this.padding}`);
  }
  build(e) {
    if (e = nt(e), e.length !== 4)
      throw new G("Input should have rank 4; Received input shape: " + JSON.stringify(e));
    let t = this.dataFormat === "channelsFirst" ? 1 : e.length - 1;
    if (e[t] == null)
      throw new G("The channel dimension of the inputs should be defined. Found `None`.");
    let n = e[t], s = this.kernelSize.concat([this.filters, n]);
    this.kernel = this.addWeight("kernel", s, "float32", this.kernelInitializer, this.kernelRegularizer, true, this.kernelConstraint), this.useBias && (this.bias = this.addWeight("bias", [this.filters], "float32", this.biasInitializer, this.biasRegularizer, true, this.biasConstraint)), this.inputSpec = [new Ft({ ndim: 4, axes: { [t]: n } })], this.built = true;
  }
  call(e, t) {
    return j(() => {
      let n = Oe(e);
      if (n.shape.length !== 4)
        throw new G(`Conv2DTranspose.call() expects input tensor to be rank-4, but received a tensor of rank-${n.shape.length}`);
      let s = n.shape, r = s[0], a, i;
      this.dataFormat === "channelsFirst" ? (a = 2, i = 3) : (a = 1, i = 2);
      let o = s[a], u = s[i], l = this.kernelSize[0], c = this.kernelSize[1], p = this.strides[0], d = this.strides[1], h = Ns(o, p, l, this.padding), f = Ns(u, d, c, this.padding), m = [r, h, f, this.filters];
      this.dataFormat !== "channelsLast" && (n = Ge(n, [0, 2, 3, 1]));
      let g = lS(n, this.kernel.read(), m, this.strides, this.padding);
      return this.dataFormat !== "channelsLast" && (g = Ge(g, [0, 3, 1, 2])), this.bias != null && (g = ks(g, this.bias.read(), this.dataFormat)), this.activation != null && (g = this.activation.apply(g)), g;
    });
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = e.slice(), n, s, r;
    this.dataFormat === "channelsFirst" ? (n = 1, s = 2, r = 3) : (n = 3, s = 1, r = 2);
    let a = this.kernelSize[0], i = this.kernelSize[1], o = this.strides[0], u = this.strides[1];
    return t[n] = this.filters, t[s] = Ns(t[s], o, a, this.padding), t[r] = Ns(t[r], u, i, this.padding), t;
  }
  getConfig() {
    let e = super.getConfig();
    return delete e.dilationRate, e;
  }
};
uy.className = "Conv2DTranspose";
re.registerClass(uy);
var ly = class extends Hp {
  constructor(e) {
    if (super(e), this.inputSpec = [new Ft({ ndim: 5 })], this.padding !== "same" && this.padding !== "valid")
      throw new G(`Conv3DTranspose currently supports only padding modes 'same' and 'valid', but received padding mode ${this.padding}`);
  }
  build(e) {
    if (e = nt(e), e.length !== 5)
      throw new G("Input should have rank 5; Received input shape: " + JSON.stringify(e));
    let t = this.dataFormat === "channelsFirst" ? 1 : e.length - 1;
    if (e[t] == null)
      throw new G("The channel dimension of the inputs should be defined. Found `None`.");
    let n = e[t], s = this.kernelSize.concat([this.filters, n]);
    this.kernel = this.addWeight("kernel", s, "float32", this.kernelInitializer, this.kernelRegularizer, true, this.kernelConstraint), this.useBias && (this.bias = this.addWeight("bias", [this.filters], "float32", this.biasInitializer, this.biasRegularizer, true, this.biasConstraint)), this.inputSpec = [new Ft({ ndim: 5, axes: { [t]: n } })], this.built = true;
  }
  call(e, t) {
    return j(() => {
      let n = Oe(e);
      if (n.shape.length !== 5)
        throw new G(`Conv3DTranspose.call() expects input tensor to be rank-4, but received a tensor of rank-${n.shape.length}`);
      let s = n.shape, r = s[0], a, i, o;
      this.dataFormat === "channelsFirst" ? (o = 2, a = 3, i = 4) : (o = 1, a = 2, i = 3);
      let u = s[o], l = s[a], c = s[i], p = this.kernelSize[0], d = this.kernelSize[1], h = this.kernelSize[2], f = this.strides[0], m = this.strides[1], g = this.strides[2], b = Ns(u, f, p, this.padding), y = Ns(l, m, d, this.padding), v = Ns(c, g, h, this.padding), x = [r, b, y, v, this.filters];
      this.dataFormat !== "channelsLast" && (n = Ge(n, [0, 2, 3, 4, 1]));
      let k = lR(n, this.kernel.read(), x, this.strides, this.padding);
      return this.dataFormat !== "channelsLast" && (k = Ge(k, [0, 4, 1, 2, 3])), this.bias !== null && (k = ks(k, this.bias.read(), this.dataFormat)), this.activation !== null && (k = this.activation.apply(k)), k;
    });
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = e.slice(), n, s, r, a;
    this.dataFormat === "channelsFirst" ? (n = 1, s = 2, r = 3, a = 4) : (n = 4, s = 1, r = 2, a = 3);
    let i = this.kernelSize[0], o = this.kernelSize[1], u = this.kernelSize[2], l = this.strides[0], c = this.strides[1], p = this.strides[2];
    return t[n] = this.filters, t[s] = Ns(t[s], l, i, this.padding), t[r] = Ns(t[r], c, o, this.padding), t[a] = Ns(t[a], p, u, this.padding), t;
  }
  getConfig() {
    let e = super.getConfig();
    return delete e.dilationRate, e;
  }
};
ly.className = "Conv3DTranspose";
re.registerClass(ly);
var WI = class extends Kl {
  constructor(e, t) {
    if (super(e, t), this.DEFAULT_DEPTHWISE_INITIALIZER = "glorotUniform", this.DEFAULT_POINTWISE_INITIALIZER = "glorotUniform", this.depthwiseKernel = null, this.pointwiseKernel = null, t.filters == null)
      throw new G("The `filters` configuration field is required by SeparableConv, but is unspecified.");
    if (t.kernelInitializer != null || t.kernelRegularizer != null || t.kernelConstraint != null)
      throw new G("Fields kernelInitializer, kernelRegularizer and kernelConstraint are invalid for SeparableConv2D. Use depthwiseInitializer, depthwiseRegularizer, depthwiseConstraint, pointwiseInitializer, pointwiseRegularizer and pointwiseConstraint instead.");
    if (t.padding != null && t.padding !== "same" && t.padding !== "valid")
      throw new G(`SeparableConv${this.rank}D supports only padding modes: 'same' and 'valid', but received ${JSON.stringify(t.padding)}`);
    this.depthMultiplier = t.depthMultiplier == null ? 1 : t.depthMultiplier, this.depthwiseInitializer = ft(t.depthwiseInitializer || this.DEFAULT_DEPTHWISE_INITIALIZER), this.depthwiseRegularizer = mt(t.depthwiseRegularizer), this.depthwiseConstraint = zt(t.depthwiseConstraint), this.pointwiseInitializer = ft(t.depthwiseInitializer || this.DEFAULT_POINTWISE_INITIALIZER), this.pointwiseRegularizer = mt(t.pointwiseRegularizer), this.pointwiseConstraint = zt(t.pointwiseConstraint);
  }
  build(e) {
    if (e = nt(e), e.length < this.rank + 2)
      throw new G(`Inputs to SeparableConv${this.rank}D should have rank ${this.rank + 2}, but received input shape: ${JSON.stringify(e)}`);
    let t = this.dataFormat === "channelsFirst" ? 1 : e.length - 1;
    if (e[t] == null || e[t] < 0)
      throw new G(`The channel dimension of the inputs should be defined, but found ${JSON.stringify(e[t])}`);
    let n = e[t], s = this.kernelSize.concat([n, this.depthMultiplier]), r = [];
    for (let i = 0; i < this.rank; ++i)
      r.push(1);
    r.push(n * this.depthMultiplier, this.filters);
    let a = true;
    this.depthwiseKernel = this.addWeight("depthwise_kernel", s, "float32", this.depthwiseInitializer, this.depthwiseRegularizer, a, this.depthwiseConstraint), this.pointwiseKernel = this.addWeight("pointwise_kernel", r, "float32", this.pointwiseInitializer, this.pointwiseRegularizer, a, this.pointwiseConstraint), this.useBias ? this.bias = this.addWeight("bias", [this.filters], "float32", this.biasInitializer, this.biasRegularizer, a, this.biasConstraint) : this.bias = null, this.inputSpec = [new Ft({ ndim: this.rank + 2, axes: { [t]: n } })], this.built = true;
  }
  call(e, t) {
    return j(() => {
      e = Oe(e);
      let n;
      if (this.rank === 1)
        throw new Fe("1D separable convolution is not implemented yet.");
      return this.rank === 2 && (this.dataFormat === "channelsFirst" && (e = Ge(e, [0, 2, 3, 1])), n = S3(e, this.depthwiseKernel.read(), this.pointwiseKernel.read(), this.strides, this.padding, this.dilationRate, "NHWC")), this.useBias && (n = ks(n, this.bias.read(), this.dataFormat)), this.activation != null && (n = this.activation.apply(n)), this.dataFormat === "channelsFirst" && (n = Ge(n, [0, 3, 1, 2])), n;
    });
  }
  getConfig() {
    let e = super.getConfig();
    return delete e.rank, delete e.kernelInitializer, delete e.kernelRegularizer, delete e.kernelConstraint, e.depthwiseInitializer = yt(this.depthwiseInitializer), e.pointwiseInitializer = yt(this.pointwiseInitializer), e.depthwiseRegularizer = it(this.depthwiseRegularizer), e.pointwiseRegularizer = it(this.pointwiseRegularizer), e.depthwiseConstraint = Pt(this.depthwiseConstraint), e.pointwiseConstraint = Pt(this.pointwiseConstraint), e;
  }
};
WI.className = "SeparableConv";
var cy = class extends WI {
  constructor(e) {
    super(2, e);
  }
};
cy.className = "SeparableConv2D";
re.registerClass(cy);
var UI = class extends Kl {
  constructor(e) {
    super(1, e), UI.verifyArgs(e), this.inputSpec = [{ ndim: 3 }];
  }
  getConfig() {
    let e = super.getConfig();
    return delete e.rank, delete e.dataFormat, e;
  }
  static verifyArgs(e) {
    if (typeof e.kernelSize != "number" && !Ab(e.kernelSize, "number", 1, 1))
      throw new G(`Conv1D expects config.kernelSize to be number or number[] with length 1, but received ${JSON.stringify(e.kernelSize)}.`);
  }
};
var dy = UI;
dy.className = "Conv1D";
re.registerClass(dy);
var py = class extends He {
  constructor(e) {
    super(e), typeof e.cropping == "number" ? this.cropping = [[e.cropping, e.cropping], [e.cropping, e.cropping]] : typeof e.cropping[0] == "number" ? this.cropping = [[e.cropping[0], e.cropping[0]], [e.cropping[1], e.cropping[1]]] : this.cropping = e.cropping, this.dataFormat = e.dataFormat === void 0 ? "channelsLast" : e.dataFormat, this.inputSpec = [{ ndim: 4 }];
  }
  computeOutputShape(e) {
    return this.dataFormat === "channelsFirst" ? [e[0], e[1], e[2] - this.cropping[0][0] - this.cropping[0][1], e[3] - this.cropping[1][0] - this.cropping[1][1]] : [e[0], e[1] - this.cropping[0][0] - this.cropping[0][1], e[2] - this.cropping[1][0] - this.cropping[1][1], e[3]];
  }
  call(e, t) {
    return j(() => {
      if (e = Oe(e), this.dataFormat === "channelsLast") {
        let n = Kc(e, this.cropping[0][0], e.shape[1] - this.cropping[0][0] - this.cropping[0][1], 2);
        return Kc(n, this.cropping[1][0], e.shape[2] - this.cropping[1][1] - this.cropping[1][0], 3);
      } else {
        let n = Kc(e, this.cropping[0][0], e.shape[2] - this.cropping[0][0] - this.cropping[0][1], 3);
        return Kc(n, this.cropping[1][0], e.shape[3] - this.cropping[1][1] - this.cropping[1][0], 4);
      }
    });
  }
  getConfig() {
    let e = { cropping: this.cropping, dataFormat: this.dataFormat }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
py.className = "Cropping2D";
re.registerClass(py);
var hy = class extends He {
  constructor(e) {
    super(e), this.DEFAULT_SIZE = [2, 2], this.inputSpec = [{ ndim: 4 }], this.size = e.size == null ? this.DEFAULT_SIZE : e.size, this.dataFormat = e.dataFormat == null ? "channelsLast" : e.dataFormat, Ct(this.dataFormat), this.interpolation = e.interpolation == null ? "nearest" : e.interpolation, rz(this.interpolation);
  }
  computeOutputShape(e) {
    if (this.dataFormat === "channelsFirst") {
      let t = e[2] == null ? null : this.size[0] * e[2], n = e[3] == null ? null : this.size[1] * e[3];
      return [e[0], e[1], t, n];
    } else {
      let t = e[1] == null ? null : this.size[0] * e[1], n = e[2] == null ? null : this.size[1] * e[2];
      return [e[0], t, n, e[3]];
    }
  }
  call(e, t) {
    return j(() => {
      let n = Oe(e), s = n.shape;
      if (this.dataFormat === "channelsFirst") {
        n = Ge(n, [0, 2, 3, 1]);
        let r = this.size[0] * s[2], a = this.size[1] * s[3], i = this.interpolation === "nearest" ? jn.resizeNearestNeighbor(n, [r, a]) : jn.resizeBilinear(n, [r, a]);
        return Ge(i, [0, 3, 1, 2]);
      } else {
        let r = this.size[0] * s[1], a = this.size[1] * s[2];
        return this.interpolation === "nearest" ? jn.resizeNearestNeighbor(n, [r, a]) : jn.resizeBilinear(n, [r, a]);
      }
    });
  }
  getConfig() {
    let e = { size: this.size, dataFormat: this.dataFormat, interpolation: this.interpolation }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
hy.className = "UpSampling2D";
re.registerClass(hy);
function sV(e, t, n = [1, 1], s = "valid", r, a) {
  return j(() => {
    r == null && (r = vs()), Ct(r);
    let i = iy(e, r);
    if (e.rank !== 4)
      throw new G(`Input for depthwiseConv2d is required to be 4-D, but is instead ${e.rank}-D`);
    if (t.rank !== 4)
      throw new G(`depthwiseKernel is required to be 4-D, but is instead ${t.rank}-D`);
    return i = xp(i, t, n, s === "same" ? "same" : "valid", "NHWC", a), r === "channelsFirst" && (i = Ge(i, [0, 3, 1, 2])), i;
  });
}
var fy = class extends oy {
  constructor(e) {
    super(2, e), this.depthwiseKernel = null, this.depthMultiplier = e.depthMultiplier == null ? 1 : e.depthMultiplier, this.depthwiseInitializer = ft(e.depthwiseInitializer || this.DEFAULT_KERNEL_INITIALIZER), this.depthwiseConstraint = zt(e.depthwiseConstraint), this.depthwiseRegularizer = mt(e.depthwiseRegularizer);
  }
  build(e) {
    if (e = nt(e), e.length < 4)
      throw new G(`Inputs to DepthwiseConv2D should have rank 4. Received input shape: ${JSON.stringify(e)}.`);
    let t = this.dataFormat === "channelsFirst" ? 1 : 3;
    if (e[t] == null || e[t] < 0)
      throw new G(`The channel dimension of the inputs to DepthwiseConv2D should be defined, but is not (${e[t]}).`);
    let n = e[t], s = [this.kernelSize[0], this.kernelSize[1], n, this.depthMultiplier];
    this.depthwiseKernel = this.addWeight("depthwise_kernel", s, null, this.depthwiseInitializer, this.depthwiseRegularizer, true, this.depthwiseConstraint), this.useBias ? this.bias = this.addWeight("bias", [n * this.depthMultiplier], null, this.biasInitializer, this.biasRegularizer, true, this.biasConstraint) : this.bias = null, this.built = true;
  }
  call(e, t) {
    return j(() => {
      e = Oe(e);
      let n = sV(e, this.depthwiseKernel.read(), this.strides, this.padding, this.dataFormat, null);
      return this.useBias && (n = ks(n, this.bias.read(), this.dataFormat)), this.activation != null && (n = this.activation.apply(n)), n;
    });
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = this.dataFormat === "channelsFirst" ? e[2] : e[1], n = this.dataFormat === "channelsFirst" ? e[3] : e[2], s = this.dataFormat === "channelsFirst" ? e[1] * this.depthMultiplier : e[3] * this.depthMultiplier, r = bs(t, this.kernelSize[0], this.padding, this.strides[0]), a = bs(n, this.kernelSize[1], this.padding, this.strides[1]);
    return this.dataFormat === "channelsFirst" ? [e[0], s, r, a] : [e[0], r, a, s];
  }
  getConfig() {
    let e = super.getConfig();
    return e.depthMultiplier = this.depthMultiplier, e.depthwiseInitializer = yt(this.depthwiseInitializer), e.depthwiseRegularizer = it(this.depthwiseRegularizer), e.depthwiseConstraint = Pt(this.depthwiseRegularizer), e;
  }
};
fy.className = "DepthwiseConv2D";
re.registerClass(fy);
function GI(e, t, n, s) {
  if (Array.isArray(e)) {
    if (t != null || n != null)
      throw new G("When inputs is an array, neither initialState or constants should be provided");
    s != null && (n = e.slice(e.length - s, e.length), e = e.slice(0, e.length - s)), e.length > 1 && (t = e.slice(1, e.length)), e = e[0];
  }
  function r(a) {
    return a == null || Array.isArray(a) ? a : [a];
  }
  return t = r(t), n = r(n), { inputs: e, initialState: t, constants: n };
}
function HI(e, t, n, s = false, r, a, i = false, o = false) {
  return j(() => {
    let u = t.shape.length;
    if (u < 3)
      throw new G(`Input should be at least 3D, but is ${u}D.`);
    let l = [1, 0].concat(ys(2, u));
    if (t = Ge(t, l), a != null)
      throw new Fe("The rnn() functoin of the deeplearn.js backend does not support constants yet.");
    i && console.warn("Backend rnn(): the unroll = true option is not applicable to the imperative deeplearn.js backend."), r != null && (r = le(le(r, "bool"), "float32"), r.rank === u - 1 && (r = Pn(r, -1)), r = Ge(r, l)), s && (t = Jn(t, 0), r != null && (r = Jn(r, 0)));
    let c = [], p, d = n, h = t.shape[0], f = Fs(t), m;
    r != null && (m = Fs(r));
    for (let b = 0; b < h; ++b) {
      let y = f[b], v = j(() => e(y, d));
      if (r == null)
        p = v[0], d = v[1];
      else {
        let x = j(() => {
          let k = m[b], I = ge(Zn(k), k), $ = ie(V(v[0], k), V(d[0], I)), R = d.map((E, P) => ie(V(v[1][P], k), V(E, I)));
          return { output: $, newStates: R };
        });
        p = x.output, d = x.newStates;
      }
      o && c.push(p);
    }
    let g;
    return o && (g = es(c, 1)), [p, g, d];
  });
}
var qI = class extends He {
  constructor(e) {
    super(e);
    let t;
    if (e.cell == null)
      throw new G("cell property is missing for the constructor of RNN.");
    if (Array.isArray(e.cell) ? t = new Kp({ cells: e.cell }) : t = e.cell, t.stateSize == null)
      throw new G("The RNN cell should have an attribute `stateSize` (tuple of integers, one integer per RNN state).");
    this.cell = t, this.returnSequences = e.returnSequences == null ? false : e.returnSequences, this.returnState = e.returnState == null ? false : e.returnState, this.goBackwards = e.goBackwards == null ? false : e.goBackwards, this._stateful = e.stateful == null ? false : e.stateful, this.unroll = e.unroll == null ? false : e.unroll, this.supportsMasking = true, this.inputSpec = [new Ft({ ndim: 3 })], this.stateSpec = null, this.states_ = null, this.numConstants = null, this.keptStates = [];
  }
  getStates() {
    if (this.states_ == null) {
      let e = Array.isArray(this.cell.stateSize) ? this.cell.stateSize.length : 1;
      return ys(0, e).map((t) => null);
    } else
      return this.states_;
  }
  setStates(e) {
    this.states_ = e;
  }
  computeOutputShape(e) {
    Cm(e) && (e = e[0]), e = e;
    let t = this.cell.stateSize;
    Array.isArray(t) || (t = [t]);
    let n = t[0], s;
    if (this.returnSequences ? s = [e[0], e[1], n] : s = [e[0], n], this.returnState) {
      let r = [];
      for (let a of t)
        r.push([e[0], a]);
      return [s].concat(r);
    } else
      return s;
  }
  computeMask(e, t) {
    return j(() => {
      Array.isArray(t) && (t = t[0]);
      let n = this.returnSequences ? t : null;
      if (this.returnState) {
        let s = this.states.map((r) => null);
        return [n].concat(s);
      } else
        return n;
    });
  }
  get states() {
    if (this.states_ == null) {
      let e = Array.isArray(this.cell.stateSize) ? this.cell.stateSize.length : 1, t = [];
      for (let n = 0; n < e; ++n)
        t.push(null);
      return t;
    } else
      return this.states_;
  }
  set states(e) {
    this.states_ = e;
  }
  build(e) {
    if (this.numConstants != null)
      throw new Fe("Constants support is not implemented in RNN yet.");
    Cm(e) && (e = e[0]), e = e;
    let n = this.stateful ? e[0] : null, s = e.slice(2);
    this.inputSpec[0] = new Ft({ shape: [n, null, ...s] });
    let r = [e[0]].concat(e.slice(2));
    this.cell.build(r);
    let a;
    if (Array.isArray(this.cell.stateSize) ? a = this.cell.stateSize : a = [this.cell.stateSize], this.stateSpec != null) {
      if (!w.arraysEqual(this.stateSpec.map((i) => i.shape[i.shape.length - 1]), a))
        throw new G(`An initialState was passed that is not compatible with cell.stateSize. Received stateSpec=${this.stateSpec}; However cell.stateSize is ${this.cell.stateSize}`);
    } else
      this.stateSpec = a.map((i) => new Ft({ shape: [null, i] }));
    this.stateful && this.resetStates();
  }
  resetStates(e, t = false) {
    j(() => {
      if (!this.stateful)
        throw new Bs("Cannot call resetStates() on an RNN Layer that is not stateful.");
      let n = this.inputSpec[0].shape[0];
      if (n == null)
        throw new G("If an RNN is stateful, it needs to know its batch size. Specify the batch size of your input tensors: \n- If using a Sequential model, specify the batch size by passing a `batchInputShape` option to your first layer.\n- If using the functional API, specify the batch size by passing a `batchShape` option to your Input layer.");
      if (this.states_ == null)
        Array.isArray(this.cell.stateSize) ? this.states_ = this.cell.stateSize.map((s) => $t([n, s])) : this.states_ = [$t([n, this.cell.stateSize])];
      else if (e == null)
        De(this.states_), this.keptStates != null && (De(this.keptStates), this.keptStates = []), Array.isArray(this.cell.stateSize) ? this.states_ = this.cell.stateSize.map((s) => $t([n, s])) : this.states_[0] = $t([n, this.cell.stateSize]);
      else {
        if (Array.isArray(e) || (e = [e]), e.length !== this.states_.length)
          throw new G(`Layer ${this.name} expects ${this.states_.length} state(s), but it received ${e.length} state value(s). Input received: ${e}`);
        t === true ? this.keptStates.push(this.states_.slice()) : De(this.states_);
        for (let s = 0; s < this.states_.length; ++s) {
          let r = e[s], a = Array.isArray(this.cell.stateSize) ? this.cell.stateSize[s] : this.cell.stateSize, i = [n, a];
          if (!w.arraysEqual(r.shape, i))
            throw new G(`State ${s} is incompatible with layer ${this.name}: expected shape=${i}, received shape=${r.shape}`);
          this.states_[s] = r;
        }
      }
      this.states_ = this.states_.map((s) => qt(s.clone()));
    });
  }
  apply(e, t) {
    let n = t == null ? null : t.initialState, s = t == null ? null : t.constants;
    t == null && (t = {});
    let r = GI(e, n, s, this.numConstants);
    e = r.inputs, n = r.initialState, s = r.constants;
    let a = [], i = [];
    if (n != null) {
      t.initialState = n, a = a.concat(n), this.stateSpec = [];
      for (let u of n)
        this.stateSpec.push(new Ft({ shape: u.shape }));
      i = i.concat(this.stateSpec);
    }
    if (s != null && (t.constants = s, a = a.concat(s), this.numConstants = s.length), a[0] instanceof $s) {
      let u = [e].concat(a), l = this.inputSpec.concat(i), c = this.inputSpec;
      this.inputSpec = l;
      let p = super.apply(u, t);
      return this.inputSpec = c, p;
    } else
      return super.apply(e, t);
  }
  call(e, t) {
    return j(() => {
      let n = t == null ? null : t.mask, s = t == null ? null : t.training, r = t == null ? null : t.initialState;
      e = Oe(e), r == null && (this.stateful ? r = this.states_ : r = this.getInitialState(e));
      let a = Array.isArray(this.cell.stateSize) ? this.cell.stateSize.length : 1;
      if (r.length !== a)
        throw new G(`RNN Layer has ${a} state(s) but was passed ${r.length} initial state(s).`);
      this.unroll && console.warn("Ignoring unroll = true for RNN layer, due to imperative backend.");
      let i = { training: s }, u = HI((h, f) => {
        let m = this.cell.call([h].concat(f), i);
        return [m[0], m.slice(1)];
      }, e, r, this.goBackwards, n, null, this.unroll, this.returnSequences), l = u[0], c = u[1], p = u[2];
      this.stateful && this.resetStates(p, s);
      let d = this.returnSequences ? c : l;
      return this.returnState ? [d].concat(p) : d;
    });
  }
  getInitialState(e) {
    return j(() => {
      let t = $t(e.shape);
      return t = ve(t, [1, 2]), t = Ul(t), Array.isArray(this.cell.stateSize) ? this.cell.stateSize.map((n) => n > 1 ? Sm(t, [1, n]) : t) : this.cell.stateSize > 1 ? [Sm(t, [1, this.cell.stateSize])] : [t];
    });
  }
  get trainableWeights() {
    return this.trainable ? this.cell.trainableWeights : [];
  }
  get nonTrainableWeights() {
    return this.trainable ? this.cell.nonTrainableWeights : this.cell.weights;
  }
  setFastWeightInitDuringBuild(e) {
    super.setFastWeightInitDuringBuild(e), this.cell != null && this.cell.setFastWeightInitDuringBuild(e);
  }
  getConfig() {
    let e = super.getConfig(), t = { returnSequences: this.returnSequences, returnState: this.returnState, goBackwards: this.goBackwards, stateful: this.stateful, unroll: this.unroll };
    this.numConstants != null && (t.numConstants = this.numConstants);
    let n = this.cell.getConfig();
    return this.getClassName() === qI.className && (t.cell = { className: this.cell.getClassName(), config: n }), { ...n, ...e, ...t };
  }
  static fromConfig(e, t, n = {}) {
    let s = t.cell, r = gs(s, n);
    return new e(Object.assign(t, { cell: r }));
  }
};
var Ar = qI;
Ar.className = "RNN";
re.registerClass(Ar);
var Xl = class extends He {
};
var qp = class extends Xl {
  constructor(e) {
    super(e), this.DEFAULT_ACTIVATION = "tanh", this.DEFAULT_KERNEL_INITIALIZER = "glorotNormal", this.DEFAULT_RECURRENT_INITIALIZER = "orthogonal", this.DEFAULT_BIAS_INITIALIZER = "zeros", this.units = e.units, Vt(this.units, "units"), this.activation = yr(e.activation == null ? this.DEFAULT_ACTIVATION : e.activation), this.useBias = e.useBias == null ? true : e.useBias, this.kernelInitializer = ft(e.kernelInitializer || this.DEFAULT_KERNEL_INITIALIZER), this.recurrentInitializer = ft(e.recurrentInitializer || this.DEFAULT_RECURRENT_INITIALIZER), this.biasInitializer = ft(e.biasInitializer || this.DEFAULT_BIAS_INITIALIZER), this.kernelRegularizer = mt(e.kernelRegularizer), this.recurrentRegularizer = mt(e.recurrentRegularizer), this.biasRegularizer = mt(e.biasRegularizer), this.kernelConstraint = zt(e.kernelConstraint), this.recurrentConstraint = zt(e.recurrentConstraint), this.biasConstraint = zt(e.biasConstraint), this.dropout = no([1, gr([0, e.dropout == null ? 0 : e.dropout])]), this.recurrentDropout = no([1, gr([0, e.recurrentDropout == null ? 0 : e.recurrentDropout])]), this.dropoutFunc = e.dropoutFunc, this.stateSize = this.units, this.dropoutMask = null, this.recurrentDropoutMask = null;
  }
  build(e) {
    e = nt(e), this.kernel = this.addWeight("kernel", [e[e.length - 1], this.units], null, this.kernelInitializer, this.kernelRegularizer, true, this.kernelConstraint), this.recurrentKernel = this.addWeight("recurrent_kernel", [this.units, this.units], null, this.recurrentInitializer, this.recurrentRegularizer, true, this.recurrentConstraint), this.useBias ? this.bias = this.addWeight("bias", [this.units], null, this.biasInitializer, this.biasRegularizer, true, this.biasConstraint) : this.bias = null, this.built = true;
  }
  call(e, t) {
    return j(() => {
      if (e = e, e.length !== 2)
        throw new G(`SimpleRNNCell expects 2 input Tensors, got ${e.length}.`);
      let n = e[1];
      e = e[0];
      let s = t.training == null ? false : t.training;
      0 < this.dropout && this.dropout < 1 && this.dropoutMask == null && (this.dropoutMask = vr({ ones: () => Zn(e), rate: this.dropout, training: s, dropoutFunc: this.dropoutFunc })), 0 < this.recurrentDropout && this.recurrentDropout < 1 && this.recurrentDropoutMask == null && (this.recurrentDropoutMask = vr({ ones: () => Zn(n), rate: this.recurrentDropout, training: s, dropoutFunc: this.dropoutFunc }));
      let r, a = this.dropoutMask, i = this.recurrentDropoutMask;
      a != null ? r = Es(V(e, a), this.kernel.read()) : r = Es(e, this.kernel.read()), this.bias != null && (r = ks(r, this.bias.read())), i != null && (n = V(n, i));
      let o = ie(r, Es(n, this.recurrentKernel.read()));
      return this.activation != null && (o = this.activation.apply(o)), [o, o];
    });
  }
  getConfig() {
    let e = super.getConfig(), t = { units: this.units, activation: br(this.activation), useBias: this.useBias, kernelInitializer: yt(this.kernelInitializer), recurrentInitializer: yt(this.recurrentInitializer), biasInitializer: yt(this.biasInitializer), kernelRegularizer: it(this.kernelRegularizer), recurrentRegularizer: it(this.recurrentRegularizer), biasRegularizer: it(this.biasRegularizer), activityRegularizer: it(this.activityRegularizer), kernelConstraint: Pt(this.kernelConstraint), recurrentConstraint: Pt(this.recurrentConstraint), biasConstraint: Pt(this.biasConstraint), dropout: this.dropout, recurrentDropout: this.recurrentDropout };
    return { ...e, ...t };
  }
};
qp.className = "SimpleRNNCell";
re.registerClass(qp);
var my = class extends Ar {
  constructor(e) {
    e.cell = new qp(e), super(e);
  }
  call(e, t) {
    return j(() => {
      this.cell.dropoutMask != null && (De(this.cell.dropoutMask), this.cell.dropoutMask = null), this.cell.recurrentDropoutMask != null && (De(this.cell.recurrentDropoutMask), this.cell.recurrentDropoutMask = null);
      let n = t == null ? null : t.mask, s = t == null ? null : t.training, r = t == null ? null : t.initialState;
      return super.call(e, { mask: n, training: s, initialState: r });
    });
  }
  static fromConfig(e, t) {
    return new e(t);
  }
};
my.className = "SimpleRNN";
re.registerClass(my);
var jp = class extends Xl {
  constructor(e) {
    if (super(e), this.DEFAULT_ACTIVATION = "tanh", this.DEFAULT_RECURRENT_ACTIVATION = "hardSigmoid", this.DEFAULT_KERNEL_INITIALIZER = "glorotNormal", this.DEFAULT_RECURRENT_INITIALIZER = "orthogonal", this.DEFAULT_BIAS_INITIALIZER = "zeros", e.resetAfter)
      throw new G("GRUCell does not support reset_after parameter set to true.");
    this.units = e.units, Vt(this.units, "units"), this.activation = yr(e.activation === void 0 ? this.DEFAULT_ACTIVATION : e.activation), this.recurrentActivation = yr(e.recurrentActivation === void 0 ? this.DEFAULT_RECURRENT_ACTIVATION : e.recurrentActivation), this.useBias = e.useBias == null ? true : e.useBias, this.kernelInitializer = ft(e.kernelInitializer || this.DEFAULT_KERNEL_INITIALIZER), this.recurrentInitializer = ft(e.recurrentInitializer || this.DEFAULT_RECURRENT_INITIALIZER), this.biasInitializer = ft(e.biasInitializer || this.DEFAULT_BIAS_INITIALIZER), this.kernelRegularizer = mt(e.kernelRegularizer), this.recurrentRegularizer = mt(e.recurrentRegularizer), this.biasRegularizer = mt(e.biasRegularizer), this.kernelConstraint = zt(e.kernelConstraint), this.recurrentConstraint = zt(e.recurrentConstraint), this.biasConstraint = zt(e.biasConstraint), this.dropout = no([1, gr([0, e.dropout == null ? 0 : e.dropout])]), this.recurrentDropout = no([1, gr([0, e.recurrentDropout == null ? 0 : e.recurrentDropout])]), this.dropoutFunc = e.dropoutFunc, this.implementation = e.implementation, this.stateSize = this.units, this.dropoutMask = null, this.recurrentDropoutMask = null;
  }
  build(e) {
    e = nt(e);
    let t = e[e.length - 1];
    this.kernel = this.addWeight("kernel", [t, this.units * 3], null, this.kernelInitializer, this.kernelRegularizer, true, this.kernelConstraint), this.recurrentKernel = this.addWeight("recurrent_kernel", [this.units, this.units * 3], null, this.recurrentInitializer, this.recurrentRegularizer, true, this.recurrentConstraint), this.useBias ? this.bias = this.addWeight("bias", [this.units * 3], null, this.biasInitializer, this.biasRegularizer, true, this.biasConstraint) : this.bias = null, this.built = true;
  }
  call(e, t) {
    return j(() => {
      if (e = e, e.length !== 2)
        throw new G(`GRUCell expects 2 input Tensors (inputs, h, c), got ${e.length}.`);
      let n = t.training == null ? false : t.training, s = e[1];
      e = e[0], 0 < this.dropout && this.dropout < 1 && this.dropoutMask == null && (this.dropoutMask = vr({ ones: () => Zn(e), rate: this.dropout, training: n, count: 3, dropoutFunc: this.dropoutFunc })), 0 < this.recurrentDropout && this.recurrentDropout < 1 && this.recurrentDropoutMask == null && (this.recurrentDropoutMask = vr({ ones: () => Zn(s), rate: this.recurrentDropout, training: n, count: 3, dropoutFunc: this.dropoutFunc }));
      let r = this.dropoutMask, a = this.recurrentDropoutMask, i, o, u;
      0 < this.dropout && this.dropout < 1 && (e = V(e, r[0]));
      let l = Es(e, this.kernel.read());
      this.useBias && (l = ks(l, this.bias.read())), 0 < this.recurrentDropout && this.recurrentDropout < 1 && (s = V(s, a[0]));
      let c = this.recurrentKernel.read(), [p, d] = Bn(c, [2 * this.units, this.units], c.rank - 1), h = Es(s, p), [f, m, g] = Bn(l, 3, l.rank - 1), [b, y] = Bn(h, 2, h.rank - 1);
      i = this.recurrentActivation.apply(ie(f, b)), o = this.recurrentActivation.apply(ie(m, y));
      let v = Es(V(o, s), d);
      u = this.activation.apply(ie(g, v));
      let x = ie(V(i, s), V(ie(1, kt(i)), u));
      return [x, x];
    });
  }
  getConfig() {
    let e = super.getConfig(), t = { units: this.units, activation: br(this.activation), recurrentActivation: br(this.recurrentActivation), useBias: this.useBias, kernelInitializer: yt(this.kernelInitializer), recurrentInitializer: yt(this.recurrentInitializer), biasInitializer: yt(this.biasInitializer), kernelRegularizer: it(this.kernelRegularizer), recurrentRegularizer: it(this.recurrentRegularizer), biasRegularizer: it(this.biasRegularizer), activityRegularizer: it(this.activityRegularizer), kernelConstraint: Pt(this.kernelConstraint), recurrentConstraint: Pt(this.recurrentConstraint), biasConstraint: Pt(this.biasConstraint), dropout: this.dropout, recurrentDropout: this.recurrentDropout, implementation: this.implementation, resetAfter: false };
    return { ...e, ...t };
  }
};
jp.className = "GRUCell";
re.registerClass(jp);
var gy = class extends Ar {
  constructor(e) {
    e.implementation === 0 && console.warn("`implementation=0` has been deprecated, and now defaults to `implementation=1`. Please update your layer call."), e.cell = new jp(e), super(e);
  }
  call(e, t) {
    return j(() => {
      this.cell.dropoutMask != null && (De(this.cell.dropoutMask), this.cell.dropoutMask = null), this.cell.recurrentDropoutMask != null && (De(this.cell.recurrentDropoutMask), this.cell.recurrentDropoutMask = null);
      let n = t == null ? null : t.mask, s = t == null ? null : t.training, r = t == null ? null : t.initialState;
      return super.call(e, { mask: n, training: s, initialState: r });
    });
  }
  static fromConfig(e, t) {
    return t.implmentation === 0 && (t.implementation = 1), new e(t);
  }
};
gy.className = "GRU";
re.registerClass(gy);
var Yl = class extends Xl {
  constructor(e) {
    super(e), this.DEFAULT_ACTIVATION = "tanh", this.DEFAULT_RECURRENT_ACTIVATION = "hardSigmoid", this.DEFAULT_KERNEL_INITIALIZER = "glorotNormal", this.DEFAULT_RECURRENT_INITIALIZER = "orthogonal", this.DEFAULT_BIAS_INITIALIZER = "zeros", this.units = e.units, Vt(this.units, "units"), this.activation = yr(e.activation === void 0 ? this.DEFAULT_ACTIVATION : e.activation), this.recurrentActivation = yr(e.recurrentActivation === void 0 ? this.DEFAULT_RECURRENT_ACTIVATION : e.recurrentActivation), this.useBias = e.useBias == null ? true : e.useBias, this.kernelInitializer = ft(e.kernelInitializer || this.DEFAULT_KERNEL_INITIALIZER), this.recurrentInitializer = ft(e.recurrentInitializer || this.DEFAULT_RECURRENT_INITIALIZER), this.biasInitializer = ft(e.biasInitializer || this.DEFAULT_BIAS_INITIALIZER), this.unitForgetBias = e.unitForgetBias, this.kernelRegularizer = mt(e.kernelRegularizer), this.recurrentRegularizer = mt(e.recurrentRegularizer), this.biasRegularizer = mt(e.biasRegularizer), this.kernelConstraint = zt(e.kernelConstraint), this.recurrentConstraint = zt(e.recurrentConstraint), this.biasConstraint = zt(e.biasConstraint), this.dropout = no([1, gr([0, e.dropout == null ? 0 : e.dropout])]), this.recurrentDropout = no([1, gr([0, e.recurrentDropout == null ? 0 : e.recurrentDropout])]), this.dropoutFunc = e.dropoutFunc, this.implementation = e.implementation, this.stateSize = [this.units, this.units], this.dropoutMask = null, this.recurrentDropoutMask = null;
  }
  build(e) {
    var t;
    e = nt(e);
    let n = e[e.length - 1];
    this.kernel = this.addWeight("kernel", [n, this.units * 4], null, this.kernelInitializer, this.kernelRegularizer, true, this.kernelConstraint), this.recurrentKernel = this.addWeight("recurrent_kernel", [this.units, this.units * 4], null, this.recurrentInitializer, this.recurrentRegularizer, true, this.recurrentConstraint);
    let s;
    if (this.useBias) {
      if (this.unitForgetBias) {
        let r = this.biasInitializer, a = this.units;
        s = new (t = class extends ns {
          apply(i, o) {
            let u = r.apply([a]), l = new Fp().apply([a]), c = r.apply([a * 2]);
            return Tx(Tx(u, l), c);
          }
        }, t.className = "CustomInit", t)();
      } else
        s = this.biasInitializer;
      this.bias = this.addWeight("bias", [this.units * 4], null, s, this.biasRegularizer, true, this.biasConstraint);
    } else
      this.bias = null;
    this.built = true;
  }
  call(e, t) {
    return j(() => {
      let n = t.training == null ? false : t.training;
      if (e = e, e.length !== 3)
        throw new G(`LSTMCell expects 3 input Tensors (inputs, h, c), got ${e.length}.`);
      let s = e[1], r = e[2];
      e = e[0], 0 < this.dropout && this.dropout < 1 && this.dropoutMask == null && (this.dropoutMask = vr({ ones: () => Zn(e), rate: this.dropout, training: n, count: 4, dropoutFunc: this.dropoutFunc })), 0 < this.recurrentDropout && this.recurrentDropout < 1 && this.recurrentDropoutMask == null && (this.recurrentDropoutMask = vr({ ones: () => Zn(s), rate: this.recurrentDropout, training: n, count: 4, dropoutFunc: this.dropoutFunc }));
      let a = this.dropoutMask, i = this.recurrentDropoutMask, o, u, l, c;
      0 < this.dropout && this.dropout < 1 && (e = V(e, a[0]));
      let p = Es(e, this.kernel.read());
      0 < this.recurrentDropout && this.recurrentDropout < 1 && (s = V(s, i[0])), p = ie(p, Es(s, this.recurrentKernel.read())), this.useBias && (p = ks(p, this.bias.read()));
      let [d, h, f, m] = Bn(p, 4, p.rank - 1);
      o = this.recurrentActivation.apply(d), u = this.recurrentActivation.apply(h), l = ie(V(u, r), V(o, this.activation.apply(f))), c = this.recurrentActivation.apply(m);
      let g = V(c, this.activation.apply(l));
      return [g, g, l];
    });
  }
  getConfig() {
    let e = super.getConfig(), t = { units: this.units, activation: br(this.activation), recurrentActivation: br(this.recurrentActivation), useBias: this.useBias, kernelInitializer: yt(this.kernelInitializer), recurrentInitializer: yt(this.recurrentInitializer), biasInitializer: yt(this.biasInitializer), unitForgetBias: this.unitForgetBias, kernelRegularizer: it(this.kernelRegularizer), recurrentRegularizer: it(this.recurrentRegularizer), biasRegularizer: it(this.biasRegularizer), activityRegularizer: it(this.activityRegularizer), kernelConstraint: Pt(this.kernelConstraint), recurrentConstraint: Pt(this.recurrentConstraint), biasConstraint: Pt(this.biasConstraint), dropout: this.dropout, recurrentDropout: this.recurrentDropout, implementation: this.implementation };
    return { ...e, ...t };
  }
};
Yl.className = "LSTMCell";
re.registerClass(Yl);
var by = class extends Ar {
  constructor(e) {
    e.implementation === 0 && console.warn("`implementation=0` has been deprecated, and now defaults to `implementation=1`. Please update your layer call."), e.cell = new Yl(e), super(e);
  }
  call(e, t) {
    return j(() => {
      this.cell.dropoutMask != null && (De(this.cell.dropoutMask), this.cell.dropoutMask = null), this.cell.recurrentDropoutMask != null && (De(this.cell.recurrentDropoutMask), this.cell.recurrentDropoutMask = null);
      let n = t == null ? null : t.mask, s = t == null ? null : t.training, r = t == null ? null : t.initialState;
      return super.call(e, { mask: n, training: s, initialState: r });
    });
  }
  static fromConfig(e, t) {
    return t.implmentation === 0 && (t.implementation = 1), new e(t);
  }
};
by.className = "LSTM";
re.registerClass(by);
var Kp = class extends Xl {
  constructor(e) {
    super(e), this.cells = e.cells;
  }
  get stateSize() {
    let e = [];
    for (let t of this.cells.slice().reverse())
      Array.isArray(t.stateSize) ? e.push(...t.stateSize) : e.push(t.stateSize);
    return e;
  }
  call(e, t) {
    return j(() => {
      e = e;
      let n = e.slice(1), s = [];
      for (let i of this.cells.slice().reverse())
        Array.isArray(i.stateSize) ? s.push(n.splice(0, i.stateSize.length)) : s.push(n.splice(0, 1));
      s.reverse();
      let r = [], a;
      for (let i = 0; i < this.cells.length; ++i) {
        let o = this.cells[i];
        n = s[i], i === 0 ? a = [e[0]].concat(n) : a = [a[0]].concat(n), a = o.call(a, t), r.push(a.slice(1));
      }
      n = [];
      for (let i of r.slice().reverse())
        n.push(...i);
      return [a[0]].concat(n);
    });
  }
  build(e) {
    Cm(e) && (e = e[0]), e = e;
    let t;
    this.cells.forEach((n, s) => {
      ta(`RNNCell_${s}`, () => {
        n.build(e), Array.isArray(n.stateSize) ? t = n.stateSize[0] : t = n.stateSize, e = [e[0], t];
      });
    }), this.built = true;
  }
  getConfig() {
    let e = super.getConfig(), t = (r) => ({ className: r.getClassName(), config: r.getConfig() }), s = { cells: this.cells.map(t) };
    return { ...e, ...s };
  }
  static fromConfig(e, t, n = {}) {
    let s = [];
    for (let r of t.cells)
      s.push(gs(r, n));
    return new e({ cells: s });
  }
  get trainableWeights() {
    if (!this.trainable)
      return [];
    let e = [];
    for (let t of this.cells)
      e.push(...t.trainableWeights);
    return e;
  }
  get nonTrainableWeights() {
    let e = [];
    for (let t of this.cells)
      e.push(...t.nonTrainableWeights);
    if (!this.trainable) {
      let t = [];
      for (let n of this.cells)
        t.push(...n.trainableWeights);
      return t.concat(e);
    }
    return e;
  }
  getWeights() {
    let e = [];
    for (let t of this.cells)
      e.push(...t.weights);
    return Nm(e);
  }
  setWeights(e) {
    let t = [];
    for (let n of this.cells) {
      let s = n.weights.length, r = e.splice(s);
      for (let a = 0; a < n.weights.length; ++a)
        t.push([n.weights[a], r[a]]);
    }
    Lb(t);
  }
};
Kp.className = "StackedRNNCells";
re.registerClass(Kp);
function vr(e) {
  let { ones: t, rate: n, training: s = false, count: r = 1, dropoutFunc: a } = e, i = () => a != null ? a(t(), n) : aI(t(), n), o = () => Hl(i, t, s);
  return !r || r <= 1 ? qt(o().clone()) : Array(r).fill(void 0).map(o).map((l) => qt(l.clone()));
}
var jI = class extends Ar {
  constructor(e) {
    if (e.unroll)
      throw new Fe("Unrolling is not possible with convolutional RNNs.");
    if (Array.isArray(e.cell))
      throw new Fe("It is not possible at the moment to stack convolutional cells.");
    super(e), this.inputSpec = [new Ft({ ndim: 5 })];
  }
  call(e, t) {
    return j(() => {
      if (this.cell.dropoutMask != null && (De(this.cell.dropoutMask), this.cell.dropoutMask = null), this.cell.recurrentDropoutMask != null && (De(this.cell.recurrentDropoutMask), this.cell.recurrentDropoutMask = null), t && t.constants)
        throw new G("ConvRNN2D cell does not support constants");
      let n = t == null ? null : t.mask, s = t == null ? null : t.training, r = t == null ? null : t.initialState;
      return super.call(e, { mask: n, training: s, initialState: r });
    });
  }
  computeOutputShape(e) {
    let t = this.computeSingleOutputShape(e);
    return this.returnSequences || (t = [t[0], ...t.slice(2)]), this.returnState && (t = [t, ...Array(2).fill([e[0], ...t.slice(-3)])]), t;
  }
  getInitialState(e) {
    return j(() => {
      let { stateSize: t } = this.cell, n = e.shape, s = this.computeSingleOutputShape(n), r = [s[0], ...s.slice(2)], a = $t(r);
      return Array.isArray(t) ? Array(t.length).fill(a) : [a];
    });
  }
  resetStates(e, t = false) {
    j(() => {
      if (!this.stateful)
        throw new Bs("Cannot call resetStates() on an RNN Layer that is not stateful.");
      let n = this.inputSpec[0].shape, s = this.computeSingleOutputShape(n), r = [s[0], ...s.slice(2)];
      if (n[0] == null)
        throw new G("If an RNN is stateful, it needs to know its batch size. Specify the batch size of your input tensors: \n- If using a Sequential model, specify the batch size by passing a `batchInputShape` option to your first layer.\n- If using the functional API, specify the batch size by passing a `batchShape` option to your Input layer.");
      if (this.getStates() == null)
        Array.isArray(this.cell.stateSize) ? this.states_ = this.cell.stateSize.map(() => $t(r)) : this.states_ = [$t(r)];
      else if (e == null)
        De(this.states_), this.keptStates != null && (De(this.keptStates), this.keptStates = []), Array.isArray(this.cell.stateSize) ? this.states_ = this.cell.stateSize.map(() => $t(r)) : this.states_[0] = $t(r);
      else {
        if (Array.isArray(e) || (e = [e]), e.length !== this.states_.length)
          throw new G(`Layer ${this.name} expects ${this.states_.length} state(s), but it received ${e.length} state value(s). Input received: ${e}`);
        t ? this.keptStates.push(this.states_.slice()) : De(this.states_);
        for (let i = 0; i < this.states_.length; ++i) {
          let o = e[i], u = r;
          if (!w.arraysEqual(o.shape, u))
            throw new G(`State ${i} is incompatible with layer ${this.name}: expected shape=${u}, received shape=${o.shape}`);
          this.states_[i] = o;
        }
      }
      this.states_ = this.states_.map((i) => qt(i.clone()));
    });
  }
  computeSingleOutputShape(e) {
    let { dataFormat: t, filters: n, kernelSize: s, padding: r, strides: a, dilationRate: i } = this.cell, o = t === "channelsFirst", u = e[o ? 3 : 2], l = e[o ? 4 : 3], c = bs(u, s[0], r, a[0], i[0]), p = bs(l, s[1], r, a[1], i[1]);
    return [...e.slice(0, 2), ...o ? [n, c, p] : [c, p, n]];
  }
};
jI.className = "ConvRNN2D";
var Xp = class extends Yl {
  constructor(e) {
    let { filters: t, kernelSize: n, strides: s, padding: r, dataFormat: a, dilationRate: i } = e;
    super({ ...e, units: t }), this.filters = t, Vt(this.filters, "filters"), this.kernelSize = Ji(n, 2, "kernelSize"), this.kernelSize.forEach((o) => Vt(o, "kernelSize")), this.strides = Ji(s || 1, 2, "strides"), this.strides.forEach((o) => Vt(o, "strides")), this.padding = r || "valid", Gn(this.padding), this.dataFormat = a || "channelsLast", Ct(this.dataFormat), this.dilationRate = Ji(i || 1, 2, "dilationRate"), this.dilationRate.forEach((o) => Vt(o, "dilationRate"));
  }
  build(e) {
    var t;
    e = nt(e);
    let n = this.dataFormat === "channelsFirst" ? 1 : e.length - 1;
    if (e[n] == null)
      throw new G(`The channel dimension of the input should be defined. Found ${e[n]}`);
    let s = e[n], r = 4, a = this.kernelSize.concat([s, this.filters * r]);
    this.kernel = this.addWeight("kernel", a, null, this.kernelInitializer, this.kernelRegularizer, true, this.kernelConstraint);
    let i = this.kernelSize.concat([this.filters, this.filters * r]);
    if (this.recurrentKernel = this.addWeight("recurrent_kernel", i, null, this.recurrentInitializer, this.recurrentRegularizer, true, this.recurrentConstraint), this.useBias) {
      let o;
      if (this.unitForgetBias) {
        let u = this.biasInitializer, l = this.filters;
        o = new (t = class extends ns {
          apply(c, p) {
            let d = u.apply([l]), h = Mn([l]), f = u.apply([l * 2]);
            return Eb([d, h, f]);
          }
        }, t.className = "CustomInit", t)();
      } else
        o = this.biasInitializer;
      this.bias = this.addWeight("bias", [this.filters * r], null, o, this.biasRegularizer, true, this.biasConstraint);
    }
    this.built = true;
  }
  call(e, t) {
    return j(() => {
      if (e.length !== 3)
        throw new G(`ConvLSTM2DCell expects 3 input Tensors (inputs, h, c), got ${e.length}.`);
      let n = t.training || false, s = e[0], r = e[1], a = e[2], i = 4;
      0 < this.dropout && this.dropout < 1 && this.dropoutMask == null && (this.dropoutMask = vr({ ones: () => Zn(s), rate: this.dropout, training: n, count: i, dropoutFunc: this.dropoutFunc }));
      let o = this.dropoutMask, u = (Z, te, J) => !te || !te[J] ? Z : V(te[J], Z), l = u(s, o, 0), c = u(s, o, 1), p = u(s, o, 2), d = u(s, o, 3);
      0 < this.recurrentDropout && this.recurrentDropout < 1 && this.recurrentDropoutMask == null && (this.recurrentDropoutMask = vr({ ones: () => Zn(r), rate: this.recurrentDropout, training: n, count: i, dropoutFunc: this.dropoutFunc }));
      let h = this.recurrentDropoutMask, f = u(r, h, 0), m = u(r, h, 1), g = u(r, h, 2), b = u(r, h, 3), y = 3, [v, x, k, I] = Bn(this.kernel.read(), i, y), [$, R, E, P] = this.useBias ? Bn(this.bias.read(), i) : [null, null, null, null];
      l = this.inputConv(l, v, $, this.padding), c = this.inputConv(c, x, R, this.padding), p = this.inputConv(p, k, E, this.padding), d = this.inputConv(d, I, P, this.padding);
      let [A, O, T, z] = Bn(this.recurrentKernel.read(), i, y);
      f = this.recurrentConv(f, A), m = this.recurrentConv(m, O), g = this.recurrentConv(g, T), b = this.recurrentConv(b, z);
      let W = this.recurrentActivation.apply(ie(l, f)), q = this.recurrentActivation.apply(ie(c, m)), X = ie(V(q, a), V(W, this.activation.apply(ie(p, g)))), Y = V(this.recurrentActivation.apply(ie(d, b)), this.activation.apply(X));
      return [Y, Y, X];
    });
  }
  getConfig() {
    let { units: e, ...t } = super.getConfig(), n = { filters: this.filters, kernelSize: this.kernelSize, padding: this.padding, dataFormat: this.dataFormat, dilationRate: this.dilationRate, strides: this.strides };
    return { ...t, ...n };
  }
  inputConv(e, t, n, s) {
    let r = da(e, t, this.strides, s || "valid", this.dataFormat === "channelsFirst" ? "NCHW" : "NHWC", this.dilationRate);
    return n ? ks(r, n, this.dataFormat) : r;
  }
  recurrentConv(e, t) {
    return da(e, t, 1, "same", this.dataFormat === "channelsFirst" ? "NCHW" : "NHWC");
  }
};
Xp.className = "ConvLSTM2DCell";
re.registerClass(Xp);
var yy = class extends jI {
  constructor(e) {
    let t = new Xp(e);
    super({ ...e, cell: t });
  }
  static fromConfig(e, t) {
    return new e(t);
  }
};
yy.className = "ConvLSTM2D";
re.registerClass(yy);
var Yp = class extends He {
  constructor(e) {
    super(e), this.rate = Math.max(Math.min(e.rate, 1), 0), this.noiseShape = e.noiseShape, this.seed = e.seed, this.supportsMasking = true;
  }
  getNoiseShape(e) {
    if (this.noiseShape == null)
      return this.noiseShape;
    let t = e.shape, n = [];
    for (let s = 0; s < this.noiseShape.length; ++s)
      n.push(this.noiseShape[s] == null ? t[s] : this.noiseShape[s]);
    return n;
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e);
      if (0 < this.rate && this.rate < 1) {
        let s = t.training == null ? false : t.training, r = this.getNoiseShape(n);
        return Hl(() => aI(n, this.rate, r, this.seed), () => n, s);
      }
      return e;
    });
  }
  getConfig() {
    let e = { rate: this.rate, noiseShape: this.noiseShape, seed: this.seed }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
  dispose() {
    return super.dispose();
  }
};
Yp.className = "Dropout";
re.registerClass(Yp);
var vy = class extends Yp {
  constructor(e) {
    super(e), this.inputSpec = [{ ndim: 3 }];
  }
  getNoiseShape(e) {
    let t = e.shape;
    return [t[0], 1, t[2]];
  }
};
vy.className = "SpatialDropout1D";
re.registerClass(vy);
var xy = class extends He {
  constructor(e) {
    if (super(e), this.activation = null, this.useBias = true, this.kernel = null, this.bias = null, this.DEFAULT_KERNEL_INITIALIZER = "glorotNormal", this.DEFAULT_BIAS_INITIALIZER = "zeros", e.batchInputShape == null && e.inputShape == null && e.inputDim != null) {
      let t = null;
      e.batchSize != null && (t = e.batchSize), this.batchInputShape = [t, e.inputDim];
    }
    this.units = e.units, Vt(this.units, "units"), this.activation = yr(e.activation), e.useBias != null && (this.useBias = e.useBias), this.kernelInitializer = ft(e.kernelInitializer || this.DEFAULT_KERNEL_INITIALIZER), this.biasInitializer = ft(e.biasInitializer || this.DEFAULT_BIAS_INITIALIZER), this.kernelConstraint = zt(e.kernelConstraint), this.biasConstraint = zt(e.biasConstraint), this.kernelRegularizer = mt(e.kernelRegularizer), this.biasRegularizer = mt(e.biasRegularizer), this.activityRegularizer = mt(e.activityRegularizer), this.supportsMasking = true, this.inputSpec = [{ minNDim: 2 }];
  }
  build(e) {
    e = nt(e);
    let t = e[e.length - 1];
    this.kernel == null && (this.kernel = this.addWeight("kernel", [t, this.units], null, this.kernelInitializer, this.kernelRegularizer, true, this.kernelConstraint), this.useBias && (this.bias = this.addWeight("bias", [this.units], null, this.biasInitializer, this.biasRegularizer, true, this.biasConstraint))), this.inputSpec = [{ minNDim: 2, axes: { [-1]: t } }], this.built = true;
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = e.slice();
    return t[t.length - 1] = this.units, t;
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e), s = ZS(this.activation.getClassName()), r;
      return s != null ? r = Es(n, this.kernel.read(), s, this.bias ? this.bias.read() : null) : (r = Es(n, this.kernel.read()), this.bias != null && (r = ks(r, this.bias.read())), this.activation != null && (r = this.activation.apply(r))), r;
    });
  }
  getConfig() {
    let e = { units: this.units, activation: br(this.activation), useBias: this.useBias, kernelInitializer: yt(this.kernelInitializer), biasInitializer: yt(this.biasInitializer), kernelRegularizer: it(this.kernelRegularizer), biasRegularizer: it(this.biasRegularizer), activityRegularizer: it(this.activityRegularizer), kernelConstraint: Pt(this.kernelConstraint), biasConstraint: Pt(this.biasConstraint) }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
xy.className = "Dense";
re.registerClass(xy);
var wy = class extends He {
  constructor(e) {
    e = e || {}, super(e), this.inputSpec = [{ minNDim: 3 }], this.dataFormat = e.dataFormat;
  }
  computeOutputShape(e) {
    e = nt(e);
    for (let t of e.slice(1))
      if (t == null)
        throw new G(`The shape of the input to "Flatten" is not fully defined (got ${e.slice(1)}). Make sure to pass a complete "input_shape" or "batch_input_shape" argument to the first layer in your model.`);
    return [e[0], cr(e, 1)];
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e);
      if (this.dataFormat === "channelsFirst" && n.rank > 1) {
        let s = [0];
        for (let r = 2; r < n.rank; ++r)
          s.push(r);
        s.push(1), n = Ge(n, s);
      }
      return cz(n);
    });
  }
  getConfig() {
    let e = {};
    this.dataFormat != null && (e.dataFormat = this.dataFormat);
    let t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
wy.className = "Flatten";
re.registerClass(wy);
var ky = class extends He {
  constructor(e) {
    super(e), this.supportsMasking = true, this.activation = yr(e.activation);
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e);
      return this.activation.apply(n);
    });
  }
  getConfig() {
    let e = { activation: br(this.activation) }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
ky.className = "Activation";
re.registerClass(ky);
var Sy = class extends He {
  constructor(e) {
    super(e), this.n = e.n, this.inputSpec = [{ ndim: 2 }];
  }
  computeOutputShape(e) {
    return [e[0], this.n, e[1]];
  }
  call(e, t) {
    return j(() => (e = Oe(e), uz(e, this.n)));
  }
  getConfig() {
    let e = { n: this.n }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
Sy.className = "RepeatVector";
re.registerClass(Sy);
var Iy = class extends He {
  constructor(e) {
    super(e), this.targetShape = e.targetShape;
    for (let t = 0; t < this.targetShape.length; ++t)
      this.isUnknown(this.targetShape[t]) && (this.targetShape[t] = null);
  }
  isUnknown(e) {
    return e < 0 || e == null;
  }
  fixUnknownDimension(e, t) {
    let n = "Total size of new array must be unchanged.", s = t.slice(), r = 1, a = null;
    for (let o = 0; o < s.length; ++o) {
      let u = s[o];
      if (this.isUnknown(u))
        if (a === null)
          a = o;
        else
          throw new G("Can only specifiy one unknown dimension.");
      else
        r *= u;
    }
    let i = cr(e);
    if (a !== null) {
      if (r === 0 || i % r !== 0)
        throw new G(n);
      s[a] = i / r;
    } else if (i !== r)
      throw new G(n);
    return s;
  }
  computeOutputShape(e) {
    let t = false;
    for (let n = 0; n < e.length; ++n)
      if (this.isUnknown(e[n])) {
        t = true;
        break;
      }
    return t ? e.slice(0, 1).concat(this.targetShape) : e.slice(0, 1).concat(this.fixUnknownDimension(e.slice(1), this.targetShape));
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e), s = n.shape, r = s.slice(0, 1).concat(this.fixUnknownDimension(s.slice(1), this.targetShape));
      return U(n, r);
    });
  }
  getConfig() {
    let e = { targetShape: this.targetShape }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
Iy.className = "Reshape";
re.registerClass(Iy);
var Cy = class extends He {
  constructor(e) {
    if (super(e), e.dims == null)
      throw new Error("Required configuration field `dims` is missing during Permute constructor call.");
    if (!Array.isArray(e.dims))
      throw new Error(`Permute constructor requires \`dims\` to be an Array, but received ${e.dims} instead.`);
    let t = ys(1, e.dims.length + 1);
    if (!w.arraysEqual(e.dims.slice().sort(), t))
      throw new Error("Invalid permutation `dims`: " + JSON.stringify(e.dims) + " `dims` must contain consecutive integers starting from 1.");
    this.dims = e.dims, this.dimsIncludingBatch = [0].concat(this.dims), this.inputSpec = [new Ft({ ndim: this.dims.length + 1 })];
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = e.slice();
    return this.dims.forEach((n, s) => {
      t[s + 1] = e[n];
    }), t;
  }
  call(e, t) {
    return Ge(Oe(e), this.dimsIncludingBatch);
  }
  getConfig() {
    let e = { dims: this.dims }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
Cy.className = "Permute";
re.registerClass(Cy);
var Ny = class extends He {
  constructor(e) {
    super(e == null ? {} : e), this.supportsMasking = true, e != null ? this.maskValue = e.maskValue == null ? 0 : e.maskValue : this.maskValue = 0;
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = super.getConfig(), t = { maskValue: this.maskValue };
    return Object.assign(t, e), t;
  }
  computeMask(e, t) {
    let n = Oe(e), s = -1;
    return ym(Ju(n, this.maskValue), s);
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e), s = -1, r = true, a = ym(Ju(n, this.maskValue), s, r);
      return V(n, le(a, n.dtype));
    });
  }
};
Ny.className = "Masking";
re.registerClass(Ny);
var Ty = class extends He {
  constructor(e) {
    if (super(e), this.embeddings = null, this.DEFAULT_EMBEDDINGS_INITIALIZER = "randomUniform", e.batchInputShape == null && e.inputShape == null) {
      let t = null;
      e.batchSize != null && (t = e.batchSize), e.inputLength == null ? this.batchInputShape = [t, null] : this.batchInputShape = [t].concat(ht(e.inputLength));
    }
    this.inputDim = e.inputDim, Vt(this.inputDim, "inputDim"), this.outputDim = e.outputDim, Vt(this.outputDim, "outputDim"), this.embeddingsInitializer = ft(e.embeddingsInitializer || this.DEFAULT_EMBEDDINGS_INITIALIZER), this.embeddingsRegularizer = mt(e.embeddingsRegularizer), this.activityRegularizer = mt(e.activityRegularizer), this.embeddingsConstraint = zt(e.embeddingsConstraint), this.maskZero = e.maskZero, this.supportsMasking = e.maskZero, this.inputLength = e.inputLength;
  }
  build(e) {
    this.embeddings = this.addWeight("embeddings", [this.inputDim, this.outputDim], this.dtype, this.embeddingsInitializer, this.embeddingsRegularizer, true, this.embeddingsConstraint), this.built = true;
  }
  warnOnIncompatibleInputShape(e) {
  }
  computeMask(e, t) {
    return j(() => this.maskZero ? (e = Oe(e), Ju(e, je(e))) : null);
  }
  computeOutputShape(e) {
    if (e = nt(e), this.inputLength == null)
      return [...e, this.outputDim];
    let t = ht(this.inputLength);
    if (t.length !== e.length - 1)
      throw new G(`"inputLength" is ${this.inputLength}, but received input shape has shape ${e}`);
    {
      let n = 0;
      for (let s = 0; s < t.length; ++s) {
        let r = t[s], a = e[s + 1];
        if (r != null && a != null && r !== a)
          throw new G(`"inputLength" is ${this.inputLength}, but received input shape has shape ${e}`);
        r == null && (t[n] = a), n++;
      }
    }
    return [e[0], ...t, this.outputDim];
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e);
      n.dtype !== "int32" && (n = Rp(n, "int32"));
      let s = rI(this.embeddings.read(), U(n, [n.size]));
      return U(s, nt(this.computeOutputShape(n.shape)));
    });
  }
  getConfig() {
    let e = { inputDim: this.inputDim, outputDim: this.outputDim, embeddingsInitializer: yt(this.embeddingsInitializer), embeddingsRegularizer: it(this.embeddingsRegularizer), activityRegularizer: it(this.activityRegularizer), embeddingsConstraint: Pt(this.embeddingsConstraint), maskZero: this.maskZero, inputLength: this.inputLength }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
Ty.className = "Embedding";
re.registerClass(Ty);
var xi = class extends He {
  constructor(e) {
    super(e || {}), this.supportsMasking = true;
  }
  mergeFunction(e) {
    throw new Fe();
  }
  computeElementwiseOpOutputShape(e, t) {
    if (e == null || t == null)
      return null;
    if (e.length < t.length)
      return this.computeElementwiseOpOutputShape(t, e);
    if (t.length === 0)
      return e;
    let n = e.slice(0, e.length - t.length);
    for (let s = 0; s < t.length; ++s) {
      let r = e[e.length - t.length + s], a = t[s];
      if (r == null || a == null || r < 0 || a < 0)
        n.push(null);
      else if (r === 1)
        n.push(a);
      else if (a === 1)
        n.push(r);
      else {
        if (r !== a)
          throw new G("Operands could not be broadcast together with shapes " + JSON.stringify(e) + " " + JSON.stringify(t));
        n.push(r);
      }
    }
    return n;
  }
  build(e) {
    if (Array.isArray(e) && !Array.isArray(e[0]) && (e = [nt(e)]), e = e, e.length < 2)
      throw new G(`A merge layer should be called on an Array of at least 2 inputs. Got ${e.length} input(s).`);
    let t = [];
    for (let r of e)
      r != null && r[0] !== null && t.push(r[0]);
    if (t = lr(t), t.length > 1)
      throw new G(`Can not merge tensors with different batch sizes. Got tensors with shapes: ${JSON.stringify(e)}.`);
    let n = e[0] == null ? null : e[0].slice(1);
    for (let r = 1; r < e.length; ++r) {
      let a = e[r] == null ? null : e[r].slice(1);
      n = this.computeElementwiseOpOutputShape(n, a);
    }
    let s = e.map((r) => r.length);
    e.indexOf(null) === -1 && lr(s).length === 1 ? this.reshapeRequired = false : this.reshapeRequired = true;
  }
  call(e, t) {
    return j(() => {
      if (e = e, this.reshapeRequired) {
        let n = [], s = e.map((r) => r.rank);
        if (s.indexOf(null) === -1) {
          let r = gr(s);
          for (let a of e) {
            let i = a.rank;
            for (let o = 0; o < r - i; ++o)
              a = Ul(a, 1);
            n.push(a);
          }
          return this.mergeFunction(n);
        } else {
          let r = false;
          for (let o of e) {
            let u = o.rank;
            if (u == null) {
              let l = o.shape, c = l[0], p = l.slice(1).concat([c]), d = U(o, [c].concat(cr(l.slice(1))));
              d = Ge(d, [1, 0]), d = U(d, p), n.push(d), r = true;
            } else if (u > 1) {
              let l = ys(1, u).concat([0]);
              n.push(Ge(o, l)), r = true;
            } else
              n.push(o);
          }
          let a = this.mergeFunction(n), i = a.rank;
          if (r) {
            if (i == null) {
              let o = a.shape, u = o.length, l = o[u - 1], c = [l].concat(o.slice(0, o.length - 1));
              a = U(Ge(U(a, [-1, l]), [1, 0]), c);
            } else if (i > 1) {
              let o = [i - 1].concat(ys(0, i - 1));
              a = Ge(a, o);
            }
          }
          return a;
        }
      } else
        return this.mergeFunction(e);
    });
  }
  computeOutputShape(e) {
    e = e;
    let t;
    e[0] == null ? t = null : t = e[0].slice(1);
    for (let s = 1; s < e.length; ++s) {
      let r = e[s] == null ? null : e[s].slice(1);
      t = this.computeElementwiseOpOutputShape(t, r);
    }
    let n = [];
    for (let s of e)
      s != null && s[0] !== null && n.push(s[0]);
    return n = lr(n), n.length === 1 ? t = n.concat(t) : t = [null].concat(t), t;
  }
  computeMask(e, t) {
    return j(() => {
      if (t == null)
        return null;
      if (!Array.isArray(t))
        throw new G("`mask` should be an Array");
      if (!Array.isArray(e))
        throw new G("`inputs` should be an Array");
      if (t.length !== e.length)
        throw new G(`The Array 'inputs' and 'mask' are expected to have the same length, but have different lengths (${e.length} vs ${t.length})`);
      if (t.every((s) => s == null))
        return null;
      t = t.map((s) => s == null ? s : Pn(s, 0));
      let n = t[0];
      for (let s = 1; s < t.length - 1; ++s)
        n = Ds(n, t[s]);
      return n;
    });
  }
};
var $y = class extends xi {
  constructor(e) {
    super(e);
  }
  mergeFunction(e) {
    return j(() => {
      let t = e[0].clone();
      for (let n = 1; n < e.length; ++n)
        t = ie(t, e[n]);
      return t;
    });
  }
};
$y.className = "Add";
re.registerClass($y);
var _y = class extends xi {
  constructor(e) {
    super(e);
  }
  mergeFunction(e) {
    return j(() => {
      let t = e[0].clone();
      for (let n = 1; n < e.length; ++n)
        t = V(t, e[n]);
      return t;
    });
  }
};
_y.className = "Multiply";
re.registerClass(_y);
var Ay = class extends xi {
  constructor(e) {
    super(e);
  }
  mergeFunction(e) {
    return j(() => {
      let t = e[0].clone();
      for (let n = 1; n < e.length; ++n)
        t = ie(t, e[n]);
      return V(1 / e.length, t);
    });
  }
};
Ay.className = "Average";
re.registerClass(Ay);
var Ey = class extends xi {
  constructor(e) {
    super(e);
  }
  mergeFunction(e) {
    return j(() => {
      let t = e[0];
      for (let n = 1; n < e.length; ++n)
        t = $r(t, e[n]);
      return t;
    });
  }
};
Ey.className = "Maximum";
re.registerClass(Ey);
var Ry = class extends xi {
  constructor(e) {
    super(e);
  }
  mergeFunction(e) {
    return j(() => {
      let t = e[0];
      for (let n = 1; n < e.length; ++n)
        t = Ip(t, e[n]);
      return t;
    });
  }
};
Ry.className = "Minimum";
re.registerClass(Ry);
var Dy = class extends xi {
  constructor(e) {
    super(e), this.DEFAULT_AXIS = -1, e == null && (e = {}), this.axis = e.axis == null ? this.DEFAULT_AXIS : e.axis, this.supportsMasking = true, this.reshapeRequired = false;
  }
  build(e) {
    if (!(Array.isArray(e) && Array.isArray(e[0])) || e.length === 1)
      throw new G("A `Concatenate` layer should be called on a list of at least 2 inputs");
    e = e;
    let t = true;
    for (let s of e)
      if (s != null) {
        t = false;
        break;
      }
    if (t)
      return;
    let n = [];
    for (let s = 0; s < e.length; ++s) {
      let r = e[s].slice();
      r.splice(this.axis, 1);
      let a = false;
      for (let i of n)
        if (w.arraysEqual(i, r)) {
          a = true;
          break;
        }
      a || n.push(r);
    }
    if (n.length > 1)
      throw new G("A `Concatenate` layer requires inputs with matching shapes except for the concat axis. Got input shapes: " + JSON.stringify(e));
  }
  mergeFunction(e) {
    return j(() => Eb(e, this.axis));
  }
  computeOutputShape(e) {
    if (!(Array.isArray(e) && Array.isArray(e[0])))
      throw new G("A `Concatenate` layer should be called on a list of inputs.");
    let t = e, n = t[0].slice(), s = this.axis < 0 ? n.length + this.axis : this.axis;
    for (let r of t.slice(1)) {
      if (n[s] == null || r[s] == null) {
        n[s] = null;
        break;
      }
      n[s] += r[s];
    }
    return n;
  }
  computeMask(e, t) {
    if (t == null)
      return null;
    if (!Array.isArray(t))
      throw new G("`mask` should be an array for Concatenate");
    if (!Array.isArray(e))
      throw new G("`inputs` should be an array for Concatenate");
    if (t.length !== e.length)
      throw new G(`Mismatch in the length of mask (${t.length}) and the legnth of inputs (${e.length})`);
    return j(() => {
      let n = true;
      if (t.forEach((a) => {
        if (a != null) {
          n = false;
          return;
        }
      }), n)
        return null;
      let s = [];
      for (let a = 0; a < e.length; ++a)
        t[a] == null ? s.push(le(Zn(e[a]), "bool")) : t[a].rank < e[a].rank ? s.push(Pn(t[a], -1)) : s.push(t[a]);
      let r = Ot(s, this.axis);
      return nS(r, -1, false);
    });
  }
  getConfig() {
    let e = { axis: this.axis }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
Dy.className = "Concatenate";
re.registerClass(Dy);
function _u(e, t) {
  for (; e < 0; )
    e += t;
  return e;
}
function rV(e, t, n) {
  if (e.shape.length > 3 || t.shape.length > 3)
    throw new Fe("batchDot is not implemented for tensors of 4D or higher rank yet");
  if (w.assert(e.shape.length >= 2, () => `batchDot requires the rank of x to be >= 2, but got ${e.shape.length}`), w.assert(e.shape.length >= 2, () => `batchDot requires the rank of y to be >= 2, but got ${t.shape.length}`), typeof n == "number" && (n = [n, n]), e.dtype === "complex64" || t.dtype === "complex64")
    throw new Fe("batchDot is not implemented for complex64-type Tensors yet.");
  let s = e.shape.length, r = t.shape.length;
  n == null && (n = [s - 1, r - 2]);
  let a = n;
  return j(() => {
    let i;
    if (s > r) {
      i = s - r;
      let u = [];
      for (let l = 0; l < i; ++l)
        u.push(1);
      t = U(t, t.shape.concat(u));
    } else if (r > s) {
      i = r - s;
      let u = [];
      for (let l = 0; l < i; ++l)
        u.push(1);
      e = U(e, e.shape.concat(u));
    } else
      i = 0;
    let o;
    if (e.shape.length === 2 && t.shape.length === 2)
      a[0] === a[1] ? o = ve(V(e, t), a[0]) : o = ve(V(Ge(e, [1, 0]), t), a[1]);
    else {
      let u = a[0] !== e.shape.length - 1, l = a[1] === t.shape.length - 1;
      o = Ve(e, t, u, l);
    }
    if (i > 0) {
      let u;
      s > r ? u = s + r - 3 : u = s - 1;
      let l = [];
      for (let c = u; c < u + i; ++c)
        l.push(c);
      o = mr(o, l);
    }
    return o.shape.length === 1 && (o = Pn(o, 1)), o;
  });
}
var Fy = class extends xi {
  constructor(e) {
    super(e), this.axes = e.axes, this.normalize = e.normalize == null ? false : e.normalize, this.supportsMasking = true, this.reshapeRequired = false;
  }
  build(e) {
    w.assert(Array.isArray(e) && e.length === 2 && Array.isArray(e[0]) && Array.isArray(e[1]), () => "A `Dot` layer should be called on a list of exactly 2 inputs.");
    let t = e[0], n = e[1];
    if (t.length > 3 || n.length > 3)
      throw new Fe("Dot layer does not support tensors of 4D or higher rank yet.");
    let s = this.interpretAxes(t, n);
    if (t[s[0]] !== n[s[1]])
      throw new G(`Dimension incompatibility: ${t[s[0]]} !== ${n[s[1]]}`);
  }
  mergeFunction(e) {
    if (e.length !== 2)
      throw new G(`A \`Dot\` layer must be called on exactly 2 inputs, but received ${e.length} input(s).`);
    let t = e[0], n = e[1], s;
    return Array.isArray(this.axes) ? s = this.axes.map((r, a) => _u(r, e[a].shape.length)) : s = [_u(this.axes, t.shape.length), _u(this.axes, n.shape.length)], this.normalize && (t = Rd(t, s[0]), n = Rd(n, s[1])), rV(t, n, s);
  }
  interpretAxes(e, t) {
    let n;
    return Array.isArray(this.axes) ? n = this.axes : n = [_u(this.axes, e.length), _u(this.axes, t.length)], n;
  }
  computeOutputShape(e) {
    w.assert(Array.isArray(e) && e.length === 2 && Array.isArray(e[0]) && Array.isArray(e[1]), () => "A `Dot` layer should be called on a list of exactly 2 inputs.");
    let t = e[0].slice(), n = e[1].slice();
    if (t.length > 3 || n.length > 3)
      throw new Fe("Dot layer does not support tensors of 4D or higher rank yet.");
    let s = this.interpretAxes(t, n);
    t.splice(s[0], 1), n.splice(s[1], 1), n.splice(0, 1);
    let r = t.concat(n);
    return r.length === 1 && r.push(1), r;
  }
  computeMask(e, t) {
    return null;
  }
  getConfig() {
    let e = { axes: this.axes, normalize: this.normalize }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
Fy.className = "Dot";
re.registerClass(Fy);
var Oy = class extends He {
  constructor(e) {
    super(e), this.supportsMasking = true, this.stddev = e.stddev;
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = super.getConfig(), t = { stddev: this.stddev };
    return Object.assign(t, e), t;
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e);
      return Hl(() => ie(Dp(n.shape, 0, this.stddev), n), () => n, t.training || false);
    });
  }
};
Oy.className = "GaussianNoise";
re.registerClass(Oy);
var Py = class extends He {
  constructor(e) {
    super(e), this.supportsMasking = true, this.rate = e.rate;
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = super.getConfig(), t = { rate: this.rate };
    return Object.assign(t, e), t;
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t);
      let n = Oe(e);
      return this.rate > 0 && this.rate < 1 ? Hl(() => {
        let r = Math.sqrt(this.rate / (1 - this.rate));
        return V(n, Dp(n.shape, 1, r));
      }, () => n, t.training || false) : n;
    });
  }
};
Py.className = "GaussianDropout";
re.registerClass(Py);
var zy = class extends He {
  constructor(e) {
    super(e), this.supportsMasking = true, this.rate = e.rate, this.noiseShape = e.noiseShape;
  }
  _getNoiseShape(e) {
    return this.noiseShape || Oe(e).shape;
  }
  computeOutputShape(e) {
    return e;
  }
  getConfig() {
    let e = super.getConfig(), t = { rate: this.rate };
    return Object.assign(t, e), t;
  }
  call(e, t) {
    return j(() => {
      if (this.rate < 1 && this.rate > 0) {
        let n = this._getNoiseShape(e);
        return Hl(() => {
          let r = Oe(e), a = 1.6732632423543772, i = 1.0507009873554805, o = -a * i, u = Zo(Vl(n), this.rate);
          u = Rp(u, "float32");
          let l = ((1 - this.rate) * (1 + this.rate * o ** 2)) ** -0.5, c = -l * o * this.rate, p = ie(V(r, u), V(ie(u, -1), o));
          return ie(V(p, l), c);
        }, () => Oe(e), t.training || false);
      }
      return e;
    });
  }
};
zy.className = "AlphaDropout";
re.registerClass(zy);
function sl(e, t, n, s, r, a = 1e-3) {
  let i;
  if (e.rank === 2)
    i = PE(e, t, n, s, r, a);
  else if (e.rank === 3)
    i = ME(e, t, n, s, r, a);
  else if (e.rank === 4)
    i = BE(e, t, n, s, r, a);
  else
    throw new Fe(`batchNormalization is not implemented for array of rank ${e.rank} yet`);
  return i;
}
function aV(e, t, n, s, r = 1e-3) {
  return j(() => {
    let a = lb(e, s), i = a.mean, o = a.variance;
    return [sl(e, i, o, n, t, r), i, o];
  });
}
function iV(e, t, n, s, r = 1e-3) {
  return j(() => {
    let a = lb(e, s), i = a.mean, o = a.variance, u = [];
    for (let f of ys(0, e.rank))
      s.indexOf(f) !== -1 ? u.push(1) : u.push(e.shape[f]);
    let l = U(i, u), c = U(o, u), p = t == null ? null : U(t, u), d = n == null ? null : U(n, u);
    return [sl(e, l, c, d, p, r), i, o];
  });
}
function oV(e, t, n, s, r = 1e-3) {
  return w.arraysEqual(s.slice().sort(), ys(0, e.rank - 1)) ? aV(e, t, n, s, r) : iV(e, t, n, s, r);
}
var My = class extends He {
  constructor(e) {
    e == null && (e = {}), super(e), this.supportsMasking = true, this.axis = e.axis == null ? -1 : e.axis, this.momentum = e.momentum == null ? 0.99 : e.momentum, this.epsilon = e.epsilon == null ? 1e-3 : e.epsilon, this.center = e.center == null ? true : e.center, this.scale = e.scale == null ? true : e.scale, this.betaInitializer = ft(e.betaInitializer || "zeros"), this.gammaInitializer = ft(e.gammaInitializer || "ones"), this.movingMeanInitializer = ft(e.movingMeanInitializer || "zeros"), this.movingVarianceInitializer = ft(e.movingVarianceInitializer || "ones"), this.betaConstraint = zt(e.betaConstraint), this.gammaConstraint = zt(e.gammaConstraint), this.betaRegularizer = mt(e.betaRegularizer), this.gammaRegularizer = mt(e.gammaRegularizer);
  }
  build(e) {
    e = nt(e);
    let t = this.axis >= 0 ? this.axis : this.axis + e.length, n = e[t];
    if (n == null)
      throw new G(`Axis ${t} of input tensor should have a defined dimension but the layer received an input with shape ${JSON.stringify(e)}.`);
    this.inputSpec = [new Ft({ ndim: e.length, axes: { [t]: n } })];
    let s = [n];
    this.scale && (this.gamma = this.addWeight("gamma", s, null, this.gammaInitializer, this.gammaRegularizer, true, this.gammaConstraint)), this.center && (this.beta = this.addWeight("beta", s, null, this.betaInitializer, this.betaRegularizer, true, this.betaConstraint)), this.movingMean = this.addWeight("moving_mean", s, null, this.movingMeanInitializer, null, false), this.movingVariance = this.addWeight("moving_variance", s, null, this.movingVarianceInitializer, null, false), this.built = true;
  }
  call(e, t) {
    return j(() => {
      let n = t.training == null ? false : t.training, s = Oe(e), r = s.shape, a = r.length, i = ys(0, a), o = this.axis >= 0 ? this.axis : this.axis + a;
      i.splice(o, 1);
      let u = ma(1, a);
      u[o] = r[o];
      let l = i.slice();
      l.sort();
      let c = !w.arraysEqual(l, ys(0, a).slice(0, a - 1)), p = () => {
        if (c) {
          let b = U(this.movingMean.read(), u), y = U(this.movingVariance.read(), u), v = this.center ? U(this.beta.read(), u) : null, x = this.scale ? U(this.gamma.read(), u) : null;
          return sl(s, b, y, v, x, this.epsilon);
        } else
          return sl(s, this.movingMean.read(), this.movingVariance.read(), this.beta == null ? null : this.beta.read(), this.gamma == null ? null : this.gamma.read(), this.epsilon);
      };
      if (!n)
        return p();
      let [d, h, f] = oV(s, this.gamma.read(), this.beta.read(), i, this.epsilon), m = (b, y, v) => {
        j(() => {
          let x = 1 - v, k = b.read(), I = V(ge(k, y), x);
          b.write(ge(k, I));
        });
      };
      return (() => {
        m(this.movingMean, h, this.momentum), m(this.movingVariance, f, this.momentum);
      })(), d;
    });
  }
  getConfig() {
    let e = { axis: this.axis, momentum: this.momentum, epsilon: this.epsilon, center: this.center, scale: this.scale, betaInitializer: yt(this.betaInitializer), gammaInitializer: yt(this.gammaInitializer), movingMeanInitializer: yt(this.movingMeanInitializer), movingVarianceInitializer: yt(this.movingVarianceInitializer), betaRegularizer: it(this.betaRegularizer), gammaRegularizer: it(this.gammaRegularizer), betaConstraint: Pt(this.betaConstraint), gammaConstraint: Pt(this.gammaConstraint) }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
My.className = "BatchNormalization";
re.registerClass(My);
var Ly = class extends He {
  constructor(e) {
    if (e == null && (e = {}), super(e), this.axis = e.axis == null ? -1 : e.axis, typeof this.axis == "number") {
      if (!Number.isInteger(this.axis))
        throw new Error(`Expected axis to be an integer, but received ${this.axis}`);
    } else if (Array.isArray(this.axis)) {
      for (let t of this.axis)
        if (!Number.isInteger(t))
          throw new Error(`Expected axis to be an array of integers, but received ${JSON.stringify(this.axis)}`);
    } else
      throw new Error(`Expected axis to be an integer or an array of integers, but received ${JSON.stringify(this.axis)}`);
    this.epsilon = e.epsilon == null ? 1e-3 : e.epsilon, this.center = e.center == null ? true : e.center, this.scale = e.scale == null ? true : e.scale, this.betaInitializer = ft(e.betaInitializer || "zeros"), this.gammaInitializer = ft(e.gammaInitializer || "ones"), this.betaRegularizer = mt(e.betaRegularizer), this.gammaRegularizer = mt(e.gammaRegularizer), this.supportsMasking = true;
  }
  build(e) {
    e = nt(e);
    let t = e.length;
    typeof this.axis == "number" && (this.axis = [this.axis]);
    for (let r = 0; r < this.axis.length; ++r)
      this.axis[r] < 0 && (this.axis[r] += t);
    for (let r of this.axis)
      if (r < 0 || r >= t)
        throw new Error(`Invalid axis: ${r}`);
    if (this.axis.length !== lr(this.axis).length)
      throw new Error(`Found duplicate axes in: ${this.axis}`);
    let n = this.axis.map((r) => e[r]), s = true;
    this.scale ? this.gamma = this.addWeight("gamma", n, "float32", this.gammaInitializer, this.gammaRegularizer, s) : this.gamma = null, this.center ? this.beta = this.addWeight("beta", n, "float32", this.betaInitializer, this.betaRegularizer, s) : this.beta = null, this.built = true;
  }
  call(e, t) {
    let n = Oe(e), s = n.shape, r = s.length;
    return j(() => {
      let { mean: i, variance: o } = lb(n, this.axis, true), u = ma(1, r);
      for (let f of this.axis)
        u[f] = s[f];
      let l = (f) => f != null && f.shape.length !== r ? U(f, u) : f, c = this.scale ? l(this.gamma.read()) : null, p = this.center ? l(this.beta.read()) : null, d = [], h = [];
      for (let f = 0; f < r; ++f)
        this.axis.indexOf(f) !== -1 ? (d.push(s[f]), h.push(1)) : (d.push(1), h.push(s[f]));
      return i = hs(i, d), o = hs(o, d), c != null && (c = hs(c, h)), p != null && (p = hs(p, h)), sl(n, i, o, p, c, this.epsilon);
    });
  }
  getConfig() {
    let e = { axis: this.axis, epsilon: this.epsilon, center: this.center, scale: this.scale, betaInitializer: yt(this.betaInitializer), gammaInitializer: yt(this.gammaInitializer), betaRegularizer: it(this.betaRegularizer), gammaRegularizer: it(this.gammaRegularizer) }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
Ly.className = "LayerNormalization";
re.registerClass(Ly);
function uV(e, t, n) {
  return j(() => {
    if (e.rank !== 4)
      throw new G(`temporalPadding expects input tensor to be 4-D, but received a ${e.rank}-D tensor.`);
    if (t == null && (t = [[1, 1], [1, 1]]), t.length !== 2 || t[0].length !== 2 || t[1].length !== 2)
      throw new G("spatial2dPadding expects `padding` to be an Array of two Arrays, each of which is an Array of two integers.");
    if (n == null && (n = vs()), n !== "channelsLast" && n !== "channelsFirst")
      throw new G(`Unknown data format: ${n}. Supported data formats are 'channelsLast' and 'channelsFirst.`);
    let s;
    return n === "channelsFirst" ? s = [[0, 0], [0, 0], t[0], t[1]] : s = [[0, 0], t[0], t[1], [0, 0]], bi(e, s);
  });
}
var By = class extends He {
  constructor(e) {
    if (e == null && (e = {}), super(e), this.dataFormat = e.dataFormat == null ? vs() : e.dataFormat, e.padding == null)
      this.padding = [[1, 1], [1, 1]];
    else if (typeof e.padding == "number")
      this.padding = [[e.padding, e.padding], [e.padding, e.padding]];
    else {
      if (e.padding = e.padding, e.padding.length !== 2)
        throw new G(`ZeroPadding2D expects padding to be a length-2 array, but received a length-${e.padding.length} array.`);
      let t, n;
      if (typeof e.padding[0] == "number")
        t = [e.padding[0], e.padding[0]], n = [e.padding[1], e.padding[1]];
      else {
        if (e.padding = e.padding, e.padding[0].length !== 2)
          throw new G(`ZeroPadding2D expects height padding to be a length-2 array, but received a length-${e.padding[0].length} array.`);
        if (t = e.padding[0], e.padding[1].length !== 2)
          throw new G(`ZeroPadding2D expects width padding to be a length-2 array, but received a length-${e.padding[1].length} array.`);
        n = e.padding[1];
      }
      this.padding = [t, n];
    }
    this.inputSpec = [new Ft({ ndim: 4 })];
  }
  computeOutputShape(e) {
    e = nt(e);
    let t, n;
    return this.dataFormat === "channelsFirst" ? (e[2] != null && e[2] >= 0 ? t = e[2] + this.padding[0][0] + this.padding[0][1] : t = null, e[3] != null && e[3] >= 0 ? n = e[3] + this.padding[1][0] + this.padding[1][1] : n = null, [e[0], e[1], t, n]) : (e[1] != null && e[1] >= 0 ? t = e[1] + this.padding[0][0] + this.padding[0][1] : t = null, e[2] != null && e[2] >= 0 ? n = e[2] + this.padding[1][0] + this.padding[1][1] : n = null, [e[0], t, n, e[3]]);
  }
  call(e, t) {
    return j(() => uV(Oe(e), this.padding, this.dataFormat));
  }
  getConfig() {
    let e = { padding: this.padding, dataFormat: this.dataFormat }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
By.className = "ZeroPadding2D";
re.registerClass(By);
function Qp(e, t, n, s, r, a) {
  return j(() => {
    Ct(r), eI(a), Gn(s), n == null && (n = [1, 1]), s == null && (s = "valid"), r == null && (r = vs()), a == null && (a = "max"), e = iy(e, r);
    let i, o = s === "same" ? "same" : "valid";
    return a === "max" ? i = ub(e, t, n, o) : i = Qg(e, t, n, o), r === "channelsFirst" && (i = Ge(i, [0, 3, 1, 2])), i;
  });
}
function KI(e, t, n, s, r, a) {
  return j(() => {
    Ct(r), eI(a), Gn(s), n == null && (n = [1, 1, 1]), s == null && (s = "valid"), r == null && (r = vs()), a == null && (a = "max"), e = LI(e, r);
    let i, o = s === "same" ? "same" : "valid";
    return a === "max" ? i = SS(e, t, n, o) : i = iS(e, t, n, o), r === "channelsFirst" && (i = Ge(i, [0, 4, 1, 2, 3])), i;
  });
}
var XI = class extends He {
  constructor(e) {
    if (e.poolSize == null && (e.poolSize = 2), super(e), typeof e.poolSize == "number")
      this.poolSize = [e.poolSize];
    else if (Array.isArray(e.poolSize) && e.poolSize.length === 1 && typeof e.poolSize[0] == "number")
      this.poolSize = e.poolSize;
    else
      throw new G(`poolSize for 1D convolutional layer must be a number or an Array of a single number, but received ${JSON.stringify(e.poolSize)}`);
    if (Vt(this.poolSize, "poolSize"), e.strides == null)
      this.strides = this.poolSize;
    else if (typeof e.strides == "number")
      this.strides = [e.strides];
    else if (Array.isArray(e.strides) && e.strides.length === 1 && typeof e.strides[0] == "number")
      this.strides = e.strides;
    else
      throw new G(`strides for 1D convolutional layer must be a number or an Array of a single number, but received ${JSON.stringify(e.strides)}`);
    Vt(this.strides, "strides"), this.padding = e.padding == null ? "valid" : e.padding, Gn(this.padding), this.inputSpec = [new Ft({ ndim: 3 })];
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = bs(e[1], this.poolSize[0], this.padding, this.strides[0]);
    return [e[0], t, e[2]];
  }
  call(e, t) {
    return j(() => {
      this.invokeCallHook(e, t), e = Ul(Oe(e), 2);
      let n = this.poolingFunction(Oe(e), [this.poolSize[0], 1], [this.strides[0], 1], this.padding, "channelsLast");
      return mr(n, [2]);
    });
  }
  getConfig() {
    let e = { poolSize: this.poolSize, padding: this.padding, strides: this.strides }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
var Vy = class extends XI {
  constructor(e) {
    super(e);
  }
  poolingFunction(e, t, n, s, r) {
    return Ct(r), Gn(s), Qp(e, t, n, s, r, "max");
  }
};
Vy.className = "MaxPooling1D";
re.registerClass(Vy);
var Wy = class extends XI {
  constructor(e) {
    super(e);
  }
  poolingFunction(e, t, n, s, r) {
    return Ct(r), Gn(s), Qp(e, t, n, s, r, "avg");
  }
};
Wy.className = "AveragePooling1D";
re.registerClass(Wy);
var YI = class extends He {
  constructor(e) {
    if (e.poolSize == null && (e.poolSize = [2, 2]), super(e), this.poolSize = Array.isArray(e.poolSize) ? e.poolSize : [e.poolSize, e.poolSize], e.strides == null)
      this.strides = this.poolSize;
    else if (Array.isArray(e.strides)) {
      if (e.strides.length !== 2)
        throw new G(`If the strides property of a 2D pooling layer is an Array, it is expected to have a length of 2, but received length ${e.strides.length}.`);
      this.strides = e.strides;
    } else
      this.strides = [e.strides, e.strides];
    Vt(this.poolSize, "poolSize"), Vt(this.strides, "strides"), this.padding = e.padding == null ? "valid" : e.padding, this.dataFormat = e.dataFormat == null ? "channelsLast" : e.dataFormat, Ct(this.dataFormat), Gn(this.padding), this.inputSpec = [new Ft({ ndim: 4 })];
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = this.dataFormat === "channelsFirst" ? e[2] : e[1], n = this.dataFormat === "channelsFirst" ? e[3] : e[2];
    return t = bs(t, this.poolSize[0], this.padding, this.strides[0]), n = bs(n, this.poolSize[1], this.padding, this.strides[1]), this.dataFormat === "channelsFirst" ? [e[0], e[1], t, n] : [e[0], t, n, e[3]];
  }
  call(e, t) {
    return j(() => (this.invokeCallHook(e, t), this.poolingFunction(Oe(e), this.poolSize, this.strides, this.padding, this.dataFormat)));
  }
  getConfig() {
    let e = { poolSize: this.poolSize, padding: this.padding, strides: this.strides, dataFormat: this.dataFormat }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
var Uy = class extends YI {
  constructor(e) {
    super(e);
  }
  poolingFunction(e, t, n, s, r) {
    return Ct(r), Gn(s), Qp(e, t, n, s, r, "max");
  }
};
Uy.className = "MaxPooling2D";
re.registerClass(Uy);
var Gy = class extends YI {
  constructor(e) {
    super(e);
  }
  poolingFunction(e, t, n, s, r) {
    return Ct(r), Gn(s), Qp(e, t, n, s, r, "avg");
  }
};
Gy.className = "AveragePooling2D";
re.registerClass(Gy);
var QI = class extends He {
  constructor(e) {
    if (e.poolSize == null && (e.poolSize = [2, 2, 2]), super(e), this.poolSize = Array.isArray(e.poolSize) ? e.poolSize : [e.poolSize, e.poolSize, e.poolSize], e.strides == null)
      this.strides = this.poolSize;
    else if (Array.isArray(e.strides)) {
      if (e.strides.length !== 3)
        throw new G(`If the strides property of a 3D pooling layer is an Array, it is expected to have a length of 3, but received length ${e.strides.length}.`);
      this.strides = e.strides;
    } else
      this.strides = [e.strides, e.strides, e.strides];
    Vt(this.poolSize, "poolSize"), Vt(this.strides, "strides"), this.padding = e.padding == null ? "valid" : e.padding, this.dataFormat = e.dataFormat == null ? "channelsLast" : e.dataFormat, Ct(this.dataFormat), Gn(this.padding), this.inputSpec = [new Ft({ ndim: 5 })];
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = this.dataFormat === "channelsFirst" ? e[2] : e[1], n = this.dataFormat === "channelsFirst" ? e[3] : e[2], s = this.dataFormat === "channelsFirst" ? e[4] : e[3];
    return t = bs(t, this.poolSize[0], this.padding, this.strides[0]), n = bs(n, this.poolSize[1], this.padding, this.strides[1]), s = bs(s, this.poolSize[2], this.padding, this.strides[2]), this.dataFormat === "channelsFirst" ? [e[0], e[1], t, n, s] : [e[0], t, n, s, e[4]];
  }
  call(e, t) {
    return j(() => (this.invokeCallHook(e, t), this.poolingFunction(Oe(e), this.poolSize, this.strides, this.padding, this.dataFormat)));
  }
  getConfig() {
    let e = { poolSize: this.poolSize, padding: this.padding, strides: this.strides, dataFormat: this.dataFormat }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
var Hy = class extends QI {
  constructor(e) {
    super(e);
  }
  poolingFunction(e, t, n, s, r) {
    return Ct(r), Gn(s), KI(e, t, n, s, r, "max");
  }
};
Hy.className = "MaxPooling3D";
re.registerClass(Hy);
var qy = class extends QI {
  constructor(e) {
    super(e);
  }
  poolingFunction(e, t, n, s, r) {
    return Ct(r), Gn(s), KI(e, t, n, s, r, "avg");
  }
};
qy.className = "AveragePooling3D";
re.registerClass(qy);
var ZI = class extends He {
  constructor(e) {
    super(e), this.inputSpec = [new Ft({ ndim: 3 })];
  }
  computeOutputShape(e) {
    return [e[0], e[2]];
  }
  call(e, t) {
    throw new Fe();
  }
};
var jy = class extends ZI {
  constructor(e) {
    super(e || {});
  }
  call(e, t) {
    return j(() => {
      let n = Oe(e);
      return It(n, 1);
    });
  }
};
jy.className = "GlobalAveragePooling1D";
re.registerClass(jy);
var Ky = class extends ZI {
  constructor(e) {
    super(e || {});
  }
  call(e, t) {
    return j(() => {
      let n = Oe(e);
      return As(n, 1);
    });
  }
};
Ky.className = "GlobalMaxPooling1D";
re.registerClass(Ky);
var JI = class extends He {
  constructor(e) {
    super(e), this.dataFormat = e.dataFormat == null ? "channelsLast" : e.dataFormat, Ct(this.dataFormat), this.inputSpec = [new Ft({ ndim: 4 })];
  }
  computeOutputShape(e) {
    return e = e, this.dataFormat === "channelsLast" ? [e[0], e[3]] : [e[0], e[1]];
  }
  call(e, t) {
    throw new Fe();
  }
  getConfig() {
    let e = { dataFormat: this.dataFormat }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
};
var Xy = class extends JI {
  call(e, t) {
    return j(() => {
      let n = Oe(e);
      return this.dataFormat === "channelsLast" ? It(n, [1, 2]) : It(n, [2, 3]);
    });
  }
};
Xy.className = "GlobalAveragePooling2D";
re.registerClass(Xy);
var Yy = class extends JI {
  call(e, t) {
    return j(() => {
      let n = Oe(e);
      return this.dataFormat === "channelsLast" ? As(n, [1, 2]) : As(n, [2, 3]);
    });
  }
};
Yy.className = "GlobalMaxPooling2D";
re.registerClass(Yy);
var e0 = class extends He {
  constructor(e) {
    super(e), this.layer = e.layer;
  }
  build(e) {
    this.built = true;
  }
  get trainable() {
    return this.layer != null ? this.layer.trainable : false;
  }
  set trainable(e) {
    this.layer != null && (this.layer.trainable = e);
  }
  get trainableWeights() {
    return this.layer.trainableWeights;
  }
  get nonTrainableWeights() {
    return this.layer.nonTrainableWeights;
  }
  get updates() {
    return this.layer._updates;
  }
  get losses() {
    return this.layer.losses;
  }
  getWeights() {
    return this.layer.getWeights();
  }
  setWeights(e) {
    this.layer.setWeights(e);
  }
  getConfig() {
    let e = { layer: { className: this.layer.getClassName(), config: this.layer.getConfig() } }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
  setFastWeightInitDuringBuild(e) {
    super.setFastWeightInitDuringBuild(e), this.layer != null && this.layer.setFastWeightInitDuringBuild(e);
  }
  static fromConfig(e, t, n = {}) {
    let s = t.layer, r = gs(s, n);
    delete t.layer;
    let a = { layer: r };
    return Object.assign(a, t), new e(a);
  }
};
var Qy = class extends e0 {
  constructor(e) {
    super(e), this.supportsMasking = true;
  }
  build(e) {
    if (e = nt(e), e.length < 3)
      throw new G(`TimeDistributed layer expects an input shape >= 3D, but received input shape ${JSON.stringify(e)}`);
    this.inputSpec = [{ shape: e }];
    let t = [e[0]].concat(e.slice(2));
    this.layer.built || (this.layer.build(t), this.layer.built = true), super.build(e);
  }
  computeOutputShape(e) {
    e = nt(e);
    let t = [e[0]].concat(e.slice(2)), n = this.layer.computeOutputShape(t), s = e[1];
    return [n[0], s].concat(n.slice(1));
  }
  call(e, t) {
    return j(() => (e = Oe(e), HI((a, i) => [Oe(this.layer.call(a, t)), []], e, [], false, null, null, false, true)[1]));
  }
};
Qy.className = "TimeDistributed";
re.registerClass(Qy);
function lV(e) {
  yi(sz, "BidirectionalMergeMode", e);
}
var cV = "concat";
var Zy = class extends e0 {
  constructor(e) {
    super(e);
    let t = e.layer.getConfig(), n = {};
    n.className = e.layer.getClassName(), n.config = t, this.forwardLayer = gs(n), t.goBackwards = t.goBackwards !== true;
    let s = {};
    if (s.className = e.layer.getClassName(), s.config = t, this.backwardLayer = gs(s), this.forwardLayer.name = "forward_" + this.forwardLayer.name, this.backwardLayer.name = "backward_" + this.backwardLayer.name, this.mergeMode = e.mergeMode === void 0 ? cV : e.mergeMode, lV(this.mergeMode), e.weights)
      throw new Fe("weights support is not implemented for Bidirectional layer yet.");
    this._stateful = e.layer.stateful, this.returnSequences = e.layer.returnSequences, this.returnState = e.layer.returnState, this.supportsMasking = true, this._trainable = true, this.inputSpec = e.layer.inputSpec, this.numConstants = null;
  }
  get trainable() {
    return this._trainable;
  }
  set trainable(e) {
    this._trainable = e, this.forwardLayer != null && (this.forwardLayer.trainable = e), this.backwardLayer != null && (this.backwardLayer.trainable = e);
  }
  getWeights() {
    return this.forwardLayer.getWeights().concat(this.backwardLayer.getWeights());
  }
  setWeights(e) {
    let t = e.length, n = Math.floor(t / 2);
    this.forwardLayer.setWeights(e.slice(0, n)), this.backwardLayer.setWeights(e.slice(n));
  }
  computeOutputShape(e) {
    let t = this.forwardLayer.computeOutputShape(e);
    Array.isArray(t) && Array.isArray(t[0]) || (t = [t]), t = t;
    let n, s, r;
    return this.returnState && (r = t.slice(1)), n = t[0], n = n, this.mergeMode === "concat" ? (n[n.length - 1] *= 2, s = [n]) : this.mergeMode == null ? s = [n, n.slice()] : s = [n], this.returnState ? this.mergeMode == null ? s.concat(r).concat(r.slice()) : [n].concat(r).concat(r.slice()) : bn(s);
  }
  apply(e, t) {
    let n = t == null ? null : t.initialState, s = t == null ? null : t.constants;
    t == null && (t = {});
    let r = GI(e, n, s, this.numConstants);
    if (e = r.inputs, n = r.initialState, s = r.constants, Array.isArray(e) && (n = e.slice(1), e = e[0]), (n == null || n.length === 0) && s == null)
      return super.apply(e, t);
    let a = [], i = [];
    if (n != null) {
      let u = n.length;
      if (u % 2 > 0)
        throw new G("When passing `initialState` to a Bidrectional RNN, the state should be an Array containing the states of the underlying RNNs.");
      t.initialState = n, a.push(...n);
      let l = n.map((c) => new Ft({ shape: c.shape }));
      this.forwardLayer.stateSpec = l.slice(0, u / 2), this.backwardLayer.stateSpec = l.slice(u / 2), i.push(...l);
    }
    if (s != null)
      throw new Fe("Support for constants in Bidirectional layers is not implemented yet.");
    let o = a[0] instanceof $s;
    for (let u of a)
      if (u instanceof $s !== o)
        throw new G("The initial state of a Bidirectional layer cannot be specified as a mix of symbolic and non-symbolic tensors");
    if (o) {
      let u = [e].concat(a), l = this.inputSpec.concat(i), c = this.inputSpec;
      this.inputSpec = l;
      let p = super.apply(u, t);
      return this.inputSpec = c, p;
    } else
      return super.apply(e, t);
  }
  call(e, t) {
    return j(() => {
      let n = t.initialState, s, r;
      if (n == null)
        s = this.forwardLayer.call(e, t), r = this.backwardLayer.call(e, t);
      else {
        let o = n.slice(0, n.length / 2), u = n.slice(n.length / 2);
        s = this.forwardLayer.call(e, Object.assign(t, { initialState: o })), r = this.backwardLayer.call(e, Object.assign(t, { initialState: u }));
      }
      let a;
      this.returnState && (Array.isArray(s) && (a = s.slice(1).concat(r.slice(1))), s = s[0], r = r[0]), this.returnSequences && (r = Jn(r, 1));
      let i;
      return this.mergeMode === "concat" ? i = Eb([s, r]) : this.mergeMode === "sum" ? i = ie(s, r) : this.mergeMode === "ave" ? i = V(0.5, ie(s, r)) : this.mergeMode === "mul" ? i = V(s, r) : this.mergeMode == null && (i = [s, r]), this.returnState ? this.mergeMode == null ? i.concat(a) : [i].concat(a) : i;
    });
  }
  resetStates(e) {
    this.forwardLayer.resetStates(), this.backwardLayer.resetStates();
  }
  build(e) {
    ta(this.forwardLayer.name, () => {
      this.forwardLayer.build(e);
    }), ta(this.backwardLayer.name, () => {
      this.backwardLayer.build(e);
    }), this.built = true;
  }
  computeMask(e, t) {
    Array.isArray(t) && (t = t[0]);
    let n;
    if (this.returnSequences ? this.mergeMode == null ? n = [t, t] : n = t : this.mergeMode == null ? n = [null, null] : n = null, this.returnState) {
      let r = this.forwardLayer.states.map((a) => null);
      return Array.isArray(n) ? n.concat(r).concat(r) : [n].concat(r).concat(r);
    } else
      return n;
  }
  get trainableWeights() {
    return this.forwardLayer.trainableWeights.concat(this.backwardLayer.trainableWeights);
  }
  get nonTrainableWeights() {
    return this.forwardLayer.nonTrainableWeights.concat(this.backwardLayer.nonTrainableWeights);
  }
  setFastWeightInitDuringBuild(e) {
    super.setFastWeightInitDuringBuild(e), this.forwardLayer != null && this.forwardLayer.setFastWeightInitDuringBuild(e), this.backwardLayer != null && this.backwardLayer.setFastWeightInitDuringBuild(e);
  }
  getConfig() {
    let e = { mergeMode: this.mergeMode }, t = super.getConfig();
    return Object.assign(e, t), e;
  }
  static fromConfig(e, t) {
    let n = gs(t.layer);
    if (delete t.layer, t.numConstants != null)
      throw new Fe("Deserialization of a Bidirectional layer with numConstants present is not supported yet.");
    let s = t;
    return s.layer = n, new e(s);
  }
};
Zy.className = "Bidirectional";
re.registerClass(Zy);
function dV(e) {
  return new tu(e);
}
function pV(e) {
  return new sy(e);
}
function hV(e) {
  return new ey(e);
}
function fV(e) {
  return new ty(e);
}
function mV(e) {
  return new ny(e);
}
function gV(e) {
  return new ay(e);
}
function bV(e) {
  return new ry(e);
}
function yV(e) {
  return new dy(e);
}
function vV(e) {
  return new Gp(e);
}
function xV(e) {
  return new uy(e);
}
function wV(e) {
  return new Hp(e);
}
function kV(e) {
  return new ly(e);
}
function SV(e) {
  return new cy(e);
}
function IV(e) {
  return new py(e);
}
function CV(e) {
  return new hy(e);
}
function NV(e) {
  return new fy(e);
}
function TV(e) {
  return new ky(e);
}
function $V(e) {
  return new xy(e);
}
function _V(e) {
  return new Yp(e);
}
function AV(e) {
  return new vy(e);
}
function EV(e) {
  return new wy(e);
}
function RV(e) {
  return new Sy(e);
}
function DV(e) {
  return new Iy(e);
}
function FV(e) {
  return new Cy(e);
}
function OV(e) {
  return new Ty(e);
}
function PV(e) {
  return new $y(e);
}
function zV(e) {
  return new Ay(e);
}
function MV(e) {
  return new Dy(e);
}
function LV(e) {
  return new Ey(e);
}
function BV(e) {
  return new Ry(e);
}
function VV(e) {
  return new _y(e);
}
function WV(e) {
  return new Fy(e);
}
function UV(e) {
  return new My(e);
}
function GV(e) {
  return new Ly(e);
}
function HV(e) {
  return new By(e);
}
function Jy(e) {
  return new Wy(e);
}
function qV(e) {
  return Jy(e);
}
function jV(e) {
  return Jy(e);
}
function ev(e) {
  return new Gy(e);
}
function KV(e) {
  return ev(e);
}
function XV(e) {
  return ev(e);
}
function tv(e) {
  return new qy(e);
}
function YV(e) {
  return tv(e);
}
function QV(e) {
  return tv(e);
}
function ZV(e) {
  return new jy(e);
}
function JV(e) {
  return new Xy(e);
}
function t0(e) {
  return new Ky(e);
}
function n0(e) {
  return new Yy(e);
}
function s0(e) {
  return new Vy(e);
}
function r0(e) {
  return new Uy(e);
}
function eW(e) {
  return new Hy(e);
}
function tW(e) {
  return new gy(e);
}
function nW(e) {
  return new jp(e);
}
function sW(e) {
  return new by(e);
}
function rW(e) {
  return new Yl(e);
}
function aW(e) {
  return new my(e);
}
function iW(e) {
  return new qp(e);
}
function oW(e) {
  return new yy(e);
}
function uW(e) {
  return new Xp(e);
}
function lW(e) {
  return new Ar(e);
}
function cW(e) {
  return new Kp(e);
}
function dW(e) {
  return new Zy(e);
}
function pW(e) {
  return new Qy(e);
}
var hW = t0;
var fW = n0;
var mW = s0;
var gW = r0;
function bW(e) {
  return new Oy(e);
}
function yW(e) {
  return new Py(e);
}
function vW(e) {
  return new zy(e);
}
function xW(e) {
  return new Ny(e);
}
var wW = {};
Ee(wW, { MAPE: () => RW, MSE: () => OW, binaryAccuracy: () => kW, binaryCrossentropy: () => SW, categoricalAccuracy: () => CW, categoricalCrossentropy: () => NW, cosineProximity: () => _W, mape: () => DW, meanAbsoluteError: () => AW, meanAbsolutePercentageError: () => EW, meanSquaredError: () => FW, mse: () => PW, precision: () => TW, recall: () => $W, sparseCategoricalAccuracy: () => IW });
function kW(e, t) {
  return jb(e, t);
}
function SW(e, t) {
  return mI(e, t);
}
function IW(e, t) {
  return gI(e, t);
}
function CW(e, t) {
  return Kb(e, t);
}
function NW(e, t) {
  return Xb(e, t);
}
function TW(e, t) {
  return fI(e, t);
}
function $W(e, t) {
  return yB(e, t);
}
function _W(e, t) {
  return qb(e, t);
}
function AW(e, t) {
  return Wp(e, t);
}
function EW(e, t) {
  return nu(e, t);
}
function RW(e, t) {
  return nu(e, t);
}
function DW(e, t) {
  return nu(e, t);
}
function FW(e, t) {
  return vi(e, t);
}
function OW(e, t) {
  return vi(e, t);
}
function PW(e, t) {
  return vi(e, t);
}
var zW = {};
Ee(zW, { modelFromJSON: () => KB });
var MW = {};
Ee(MW, { l1: () => BW, l1l2: () => LW, l2: () => VW });
function LW(e) {
  return new jl(e);
}
function BW(e) {
  return JB(e);
}
function VW(e) {
  return eV(e);
}
var WW = class extends so {
  constructor() {
    super(...arguments), this.model = null;
  }
  setModel(e) {
    if (!(e instanceof dr))
      throw new Error("model must be a LayersModel, not some other Container");
    this.model = e;
  }
};
function Yc(e, t) {
  return e < t;
}
function Yx(e, t) {
  return e > t;
}
var UW = class extends WW {
  constructor(e) {
    if (super(), e == null && (e = {}), e.restoreBestWeights)
      throw new Fe("restoreBestWeights = True is not implemented in EarlyStopping yet.");
    this.monitor = e.monitor || "val_loss", this.minDelta = Math.abs(e.minDelta || 0), this.patience = e.patience || 0, this.verbose = e.verbose || 0, this.mode = e.mode || "auto", this.baseline = e.baseline, ["auto", "min", "max"].indexOf(this.mode) === -1 && (console.warn(`EarlyStopping mode '${this.mode}' is invalid. Falling back to mode 'auto'.`), this.mode = "auto"), this.mode === "min" ? this.monitorFunc = Yc : this.mode === "max" ? this.monitorFunc = Yx : this.monitor.indexOf("acc") !== -1 ? this.monitorFunc = Yx : this.monitorFunc = Yc, this.monitorFunc === Yc && (this.minDelta *= -1);
  }
  async onTrainBegin(e) {
    this.wait = 0, this.stoppedEpoch = 0, this.baseline != null ? this.best = this.baseline : this.best = this.monitorFunc === Yc ? 1 / 0 : -1 / 0;
  }
  async onEpochEnd(e, t) {
    await sr(t);
    let n = this.getMonitorValue(t);
    n != null && (this.monitorFunc(n - this.minDelta, this.best) ? (this.best = n, this.wait = 0) : (this.wait++, this.wait >= this.patience && (this.stoppedEpoch = e, this.model.stopTraining = true)));
  }
  async onTrainEnd(e) {
    this.stoppedEpoch > 0 && this.verbose && console.log(`Epoch ${this.stoppedEpoch}: early stopping.`);
  }
  getMonitorValue(e) {
    e == null && (e = {});
    let t = e[this.monitor];
    return t == null && console.warn(`Metric for EarlyStopping ${this.monitor} is not available. Available metrics are: ${Object.keys(e)}`), t;
  }
};
function GW(e) {
  return new UW(e);
}
var uhe = { earlyStopping: GW };
var HW = K();
HW.registerFlag("KEEP_INTERMEDIATE_TENSORS", () => false, (e) => {
  e && console.warn("Keep intermediate tensors is ON. This will print the values of all intermediate tensors during model inference. Not all models support this mode. For details, check e2e/benchmarks/ model_config.js. This significantly impacts performance.");
});
var a0 = ((e) => (e[e.DT_INVALID = 0] = "DT_INVALID", e[e.DT_FLOAT = 1] = "DT_FLOAT", e[e.DT_DOUBLE = 2] = "DT_DOUBLE", e[e.DT_INT32 = 3] = "DT_INT32", e[e.DT_UINT8 = 4] = "DT_UINT8", e[e.DT_INT16 = 5] = "DT_INT16", e[e.DT_INT8 = 6] = "DT_INT8", e[e.DT_STRING = 7] = "DT_STRING", e[e.DT_COMPLEX64 = 8] = "DT_COMPLEX64", e[e.DT_INT64 = 9] = "DT_INT64", e[e.DT_BOOL = 10] = "DT_BOOL", e[e.DT_QINT8 = 11] = "DT_QINT8", e[e.DT_QUINT8 = 12] = "DT_QUINT8", e[e.DT_QINT32 = 13] = "DT_QINT32", e[e.DT_BFLOAT16 = 14] = "DT_BFLOAT16", e[e.DT_QINT16 = 15] = "DT_QINT16", e[e.DT_QUINT16 = 16] = "DT_QUINT16", e[e.DT_UINT16 = 17] = "DT_UINT16", e[e.DT_COMPLEX128 = 18] = "DT_COMPLEX128", e[e.DT_HALF = 19] = "DT_HALF", e[e.DT_RESOURCE = 20] = "DT_RESOURCE", e[e.DT_VARIANT = 21] = "DT_VARIANT", e[e.DT_UINT32 = 22] = "DT_UINT32", e[e.DT_UINT64 = 23] = "DT_UINT64", e[e.DT_FLOAT_REF = 101] = "DT_FLOAT_REF", e[e.DT_DOUBLE_REF = 102] = "DT_DOUBLE_REF", e[e.DT_INT32_REF = 103] = "DT_INT32_REF", e[e.DT_UINT8_REF = 104] = "DT_UINT8_REF", e[e.DT_INT16_REF = 105] = "DT_INT16_REF", e[e.DT_INT8_REF = 106] = "DT_INT8_REF", e[e.DT_STRING_REF = 107] = "DT_STRING_REF", e[e.DT_COMPLEX64_REF = 108] = "DT_COMPLEX64_REF", e[e.DT_INT64_REF = 109] = "DT_INT64_REF", e[e.DT_BOOL_REF = 110] = "DT_BOOL_REF", e[e.DT_QINT8_REF = 111] = "DT_QINT8_REF", e[e.DT_QUINT8_REF = 112] = "DT_QUINT8_REF", e[e.DT_QINT32_REF = 113] = "DT_QINT32_REF", e[e.DT_BFLOAT16_REF = 114] = "DT_BFLOAT16_REF", e[e.DT_QINT16_REF = 115] = "DT_QINT16_REF", e[e.DT_QUINT16_REF = 116] = "DT_QUINT16_REF", e[e.DT_UINT16_REF = 117] = "DT_UINT16_REF", e[e.DT_COMPLEX128_REF = 118] = "DT_COMPLEX128_REF", e[e.DT_HALF_REF = 119] = "DT_HALF_REF", e[e.DT_RESOURCE_REF = 120] = "DT_RESOURCE_REF", e[e.DT_VARIANT_REF = 121] = "DT_VARIANT_REF", e[e.DT_UINT32_REF = 122] = "DT_UINT32_REF", e[e.DT_UINT64_REF = 123] = "DT_UINT64_REF", e))(a0 || {});
var Qx;
((e) => {
  let t;
  ((n) => {
    n[n.LEGACY = 0] = "LEGACY", n[n.V1 = 1] = "V1", n[n.V2 = 2] = "V2";
  })(t = e.CheckpointFormatVersion || (e.CheckpointFormatVersion = {}));
})(Qx || (Qx = {}));
var nv = {};
function lhe(e, t) {
  let n = { tfOpName: e, category: "custom", inputs: [], attrs: [], customExecutor: t };
  nv[e] = n;
}
function i0(e) {
  return nv[e];
}
function che(e) {
  delete nv[e];
}
function S(e, t, n, s, r) {
  let a = t.inputParams[e];
  if (a && a.inputIndexStart !== void 0) {
    let o = a.inputIndexStart, u = a.inputIndexEnd === 0 ? void 0 : a.inputIndexEnd === void 0 ? o + 1 : a.inputIndexEnd;
    if (a.type === "tensor")
      return un(t.inputNames[a.inputIndexStart], n, s, r);
    if (a.type === "tensors")
      return t.inputNames.slice(o, u).map((d) => un(d, n, s, r));
    let l = un(t.inputNames.slice(o)[0], n, s, r), c = l.dataSync();
    return a.type === "number" ? c[0] : w.toNestedArray(l.shape, c);
  }
  let i = t.attrParams[e];
  return i && i.value;
}
function un(e, t, n, s) {
  let [r, a] = _n(e);
  if (s != null) {
    let o = s.getHashTableHandleByName(r);
    if (o != null)
      return o;
  }
  let i = n.currentContextIds.find((o) => !!t[zd(r, o)]);
  return i !== void 0 ? t[zd(r, i)][a] : void 0;
}
function qW(e, t, n) {
  return t[zd(e, n.currentContextId)];
}
function Ts(e, t) {
  let [n, s, r] = _n(e);
  return [zd(n, t && t.currentContextId), s, r];
}
function zd(e, t) {
  return t ? `${e}-${t}` : e;
}
function _n(e) {
  let t = e.split(":");
  if (t.length === 1)
    return [e, 0, void 0];
  let n = t[0], s = t.length === 3 ? t[1] : void 0, r = Number(t[t.length - 1]);
  return [n, r, s];
}
function id(e, t, n) {
  let s = S("pad", e, t, n);
  if (s === "explicit") {
    s = S("explicitPaddings", e, t, n);
    let r = [[0, 0], [0, 0], [0, 0], [0, 0]];
    for (let a = 0; a < 4; a++)
      r[a][0] = s[a * 2], r[a][1] = s[a * 2 + 1];
    return r;
  }
  return s;
}
function Ws(e) {
  return e.kept ? e : ur(e);
}
var o0 = {};
Ee(o0, { json: () => jW });
var jW = [{ tfOpName: "Add", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "AddV2", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "AddN", category: "arithmetic", inputs: [{ start: 0, end: 0, name: "tensors", type: "tensors" }] }, { tfOpName: "BiasAdd", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }] }, { tfOpName: "Sub", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "RealDiv", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Div", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "DivNoNan", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "FloorDiv", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Mul", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Maximum", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Minimum", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Pow", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "SquaredDifference", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Mod", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "FloorMod", category: "arithmetic", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }];
var u0 = {};
Ee(u0, { json: () => KW });
var KW = [{ tfOpName: "Abs", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Acos", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Asin", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Atan", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Atan2", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "y", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Ceil", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "ClipByValue", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "clipValueMin", type: "number" }, { start: 2, name: "clipValueMax", type: "number" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Complex", category: "basic_math", inputs: [{ start: 0, name: "real", type: "tensor" }, { start: 1, name: "imag", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "ComplexAbs", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Cos", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Cosh", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Elu", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Exp", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Floor", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Log", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Imag", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "Tout", name: "outputType", type: "dtype", notSupported: true }] }, { tfOpName: "Neg", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Real", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "Tout", name: "outputType", type: "dtype", notSupported: true }] }, { tfOpName: "Prelu", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "alpha", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Relu", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Relu6", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Selu", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Sigmoid", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Sin", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Sinh", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Sqrt", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Rsqrt", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Square", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Tan", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Tanh", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Sign", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Round", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Expm1", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Log1p", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Reciprocal", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Softplus", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Asinh", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Acosh", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Atanh", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Erf", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Prod", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axes", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool", notSupported: true }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "LeakyRelu", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "alpha", name: "alpha", type: "number", defaultValue: 0.2 }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "IsNan", category: "basic_math", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }];
var l0 = {};
Ee(l0, { json: () => XW });
var XW = [{ tfOpName: "EmptyTensorList", category: "control", inputs: [{ start: 0, name: "elementShape", type: "shape" }, { start: 1, name: "maxNumElements", type: "number" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "LoopCond", category: "control", inputs: [{ start: 0, name: "pred", type: "tensor" }] }, { tfOpName: "Switch", category: "control", inputs: [{ start: 0, name: "data", type: "tensor" }, { start: 1, name: "pred", type: "tensor" }] }, { tfOpName: "Merge", category: "control", inputs: [{ start: 0, end: 0, name: "tensors", type: "tensors" }] }, { tfOpName: "Enter", category: "control", inputs: [{ start: 0, name: "tensor", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "frame_name", name: "frameName", type: "string" }, { tfName: "is_constant", name: "isConstant", type: "bool" }] }, { tfOpName: "Exit", category: "control", inputs: [{ start: 0, name: "tensor", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "NextIteration", category: "control", inputs: [{ start: 0, name: "tensor", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "TensorArrayV3", category: "control", inputs: [{ start: 0, name: "size", type: "number" }], attrs: [{ tfName: "dtype", name: "dtype", type: "dtype" }, { tfName: "element_shape", name: "elementShape", type: "shape" }, { tfName: "dynamic_size", name: "dynamicSize", type: "bool" }, { tfName: "clear_after_read", name: "clearAfterRead", type: "bool" }, { tfName: "identical_element_shapes", name: "identicalElementShapes", type: "bool" }, { tfName: "tensor_array_name", name: "name", type: "string" }] }, { tfOpName: "TensorArrayWriteV3", category: "control", inputs: [{ start: 0, name: "tensorArrayId", type: "tensor" }, { start: 1, name: "index", type: "number" }, { start: 2, name: "tensor", type: "tensor" }, { start: 3, name: "flowIn", type: "number" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "TensorArrayReadV3", category: "control", inputs: [{ start: 0, name: "tensorArrayId", type: "tensor" }, { start: 1, name: "index", type: "number" }, { start: 2, name: "flowIn", type: "number" }], attrs: [{ tfName: "dtype", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "TensorArrayGatherV3", category: "control", inputs: [{ start: 0, name: "tensorArrayId", type: "tensor" }, { start: 1, name: "indices", type: "number[]" }, { start: 2, name: "flowIn", type: "number" }], attrs: [{ tfName: "dtype", name: "dtype", type: "dtype" }, { tfName: "element_shape", name: "elementShape", type: "shape" }] }, { tfOpName: "TensorArrayScatterV3", category: "control", inputs: [{ start: 0, name: "tensorArrayId", type: "tensor" }, { start: 1, name: "indices", type: "number[]" }, { start: 2, name: "tensor", type: "tensor" }, { start: 3, name: "flowIn", type: "number" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype" }] }, { tfOpName: "TensorArrayConcatV3", category: "control", inputs: [{ start: 0, name: "tensorArrayId", type: "tensor" }, { start: 1, name: "flowIn", type: "number" }], attrs: [{ tfName: "dtype", name: "dtype", type: "dtype" }, { tfName: "element_shape_except0", name: "elementShapeExcept0", type: "shape", notSupported: true }] }, { tfOpName: "TensorArraySplitV3", category: "control", inputs: [{ start: 0, name: "tensorArrayId", type: "tensor" }, { start: 1, name: "tensor", type: "tensor" }, { start: 2, name: "lengths", type: "number[]" }, { start: 3, name: "flowIn", type: "number" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype" }] }, { tfOpName: "TensorArraySizeV3", category: "control", inputs: [{ start: 0, name: "tensorArrayId", type: "tensor" }, { start: 1, name: "flowIn", type: "number" }] }, { tfOpName: "TensorArrayCloseV3", category: "control", inputs: [{ start: 0, name: "tensorArrayId", type: "tensor" }] }, { tfOpName: "StatelessIf", category: "control", inputs: [{ start: 0, name: "cond", type: "tensor" }, { start: 1, end: 0, name: "args", type: "tensors" }], attrs: [{ tfName: "then_branch", name: "thenBranch", type: "func" }, { tfName: "else_branch", name: "elseBranch", type: "func" }] }, { tfOpName: "If", category: "control", inputs: [{ start: 0, name: "cond", type: "tensor" }, { start: 1, end: 0, name: "args", type: "tensors" }], attrs: [{ tfName: "then_branch", name: "thenBranch", type: "func" }, { tfName: "else_branch", name: "elseBranch", type: "func" }] }, { tfOpName: "StatelessWhile", category: "control", inputs: [{ start: 0, end: 0, name: "args", type: "tensors" }], attrs: [{ tfName: "cond", name: "cond", type: "func" }, { tfName: "body", name: "body", type: "func" }] }, { tfOpName: "While", category: "control", inputs: [{ start: 0, end: 0, name: "args", type: "tensors" }], attrs: [{ tfName: "cond", name: "cond", type: "func" }, { tfName: "body", name: "body", type: "func" }] }, { tfOpName: "TensorListScatter", category: "control", inputs: [{ start: 0, name: "tensor", type: "tensor" }, { start: 1, name: "indices", type: "number[]" }, { start: 2, name: "elementShape", type: "shape" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListScatterV2", category: "control", inputs: [{ start: 0, name: "tensor", type: "tensor" }, { start: 1, name: "indices", type: "number[]" }, { start: 2, name: "elementShape", type: "shape" }, { start: 3, name: "numElements", type: "number" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListGather", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }, { start: 1, name: "indices", type: "number[]" }, { start: 2, name: "elementShape", type: "shape" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListGetItem", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }, { start: 1, name: "index", type: "number" }, { start: 2, name: "elementShape", type: "shape" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListSetItem", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }, { start: 1, name: "index", type: "number" }, { start: 2, name: "tensor", type: "tensor" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListReserve", category: "control", inputs: [{ start: 0, name: "elementShape", type: "shape" }, { start: 1, name: "numElements", type: "number" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListFromTensor", category: "control", inputs: [{ start: 0, name: "tensor", type: "tensor" }, { start: 1, name: "elementShape", type: "shape" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListStack", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }, { start: 1, name: "elementShape", type: "shape" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }, { tfName: "num_elements", name: "numElements", type: "dtype" }] }, { tfOpName: "TensorListSplit", category: "control", inputs: [{ start: 0, name: "tensor", type: "tensor" }, { start: 1, name: "elementShape", type: "shape" }, { start: 2, name: "lengths", type: "number[]" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListConcat", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }], attrs: [{ tfName: "element_shape", name: "elementShape", type: "shape" }, { tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListConcatV2", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }], attrs: [{ tfName: "element_shape", name: "elementShape", type: "shape" }, { tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListPopBack", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }, { start: 1, name: "elementShape", type: "shape" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListPushBack", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }, { start: 1, name: "tensor", type: "tensor" }], attrs: [{ tfName: "element_dtype", name: "elementDType", type: "dtype" }] }, { tfOpName: "TensorListLength", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }] }, { tfOpName: "TensorListResize", category: "control", inputs: [{ start: 0, name: "tensorListId", type: "tensor" }, { start: 1, name: "size", type: "number" }] }];
var c0 = {};
Ee(c0, { json: () => YW });
var YW = [{ tfOpName: "AvgPool", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }, { tfName: "ksize", name: "kernelSize", type: "number[]" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "MaxPool", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }, { tfName: "ksize", name: "kernelSize", type: "number[]" }, { tfName: "explicit_paddings", name: "explicitPaddings", type: "number[]", defaultValue: [], notSupported: true }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "MaxPoolWithArgmax", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "ksize", name: "kernelSize", type: "number[]" }, { tfName: "include_batch_in_index", name: "includeBatchInIndex", type: "bool" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "AvgPool3D", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }, { tfName: "ksize", name: "kernelSize", type: "number[]" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "MaxPool3D", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }, { tfName: "ksize", name: "kernelSize", type: "number[]" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Conv1D", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }], attrs: [{ tfName: "stride", name: "stride", type: "number" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", defaultValue: "NWC" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "dilation", name: "dilation", type: "number", defaultValue: 1 }] }, { tfOpName: "Conv2D", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "useCudnnOnGpu", name: "useCudnnOnGpu", type: "bool" }, { tfName: "data_format", name: "dataFormat", type: "string", defaultValue: "NHWC" }, { tfName: "explicit_paddings", name: "explicitPaddings", type: "number[]", defaultValue: [] }, { tfName: "dilations", name: "dilations", type: "number[]" }] }, { tfOpName: "_FusedConv2D", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }, { start: 2, end: 0, name: "args", type: "tensors" }], attrs: [{ tfName: "num_args", name: "numArgs", type: "number" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "explicit_paddings", name: "explicitPaddings", type: "number[]", defaultValue: [] }, { tfName: "use_cudnn_on_gpu", name: "useCudnnOnGpu", type: "bool", defaultValue: true }, { tfName: "data_format", name: "dataFormat", type: "string", defaultValue: "NHWC" }, { tfName: "dilations", name: "dilations", type: "number[]", defaultValue: [1, 1, 1, 1] }, { tfName: "fused_ops", name: "fusedOps", type: "string[]", defaultValue: [] }, { tfName: "epsilon", name: "epsilon", type: "number", defaultValue: 1e-4 }, { tfName: "leakyrelu_alpha", name: "leakyreluAlpha", type: "number" }] }, { tfOpName: "Conv2DBackpropInput", category: "convolution", inputs: [{ start: 2, name: "x", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }, { start: 0, name: "outputShape", type: "number[]" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }, { tfName: "explicit_paddings", name: "explicitPaddings", type: "number[]", defaultValue: [] }, { tfName: "dilations", name: "dilations", type: "number[]", notSupported: true }] }, { tfOpName: "DepthwiseConv2d", category: "convolution", inputs: [{ start: 0, name: "input", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", defaultValue: "NHWC" }, { tfName: "explicit_paddings", name: "explicitPaddings", type: "number[]", defaultValue: [] }, { tfName: "dilations", name: "dilations", type: "number[]" }] }, { tfOpName: "DepthwiseConv2dNative", category: "convolution", inputs: [{ start: 0, name: "input", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", defaultValue: "NHWC" }, { tfName: "explicit_paddings", name: "explicitPaddings", type: "number[]", defaultValue: [] }, { tfName: "dilations", name: "dilations", type: "number[]" }] }, { tfOpName: "FusedDepthwiseConv2dNative", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }, { start: 2, end: 0, name: "args", type: "tensors" }], attrs: [{ tfName: "num_args", name: "numArgs", type: "number" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", defaultValue: "NHWC" }, { tfName: "dilations", name: "dilations", type: "number[]", defaultValue: [1, 1, 1, 1] }, { tfName: "fused_ops", name: "fusedOps", type: "string[]", defaultValue: [] }, { tfName: "explicit_paddings", name: "explicitPaddings", type: "number[]", defaultValue: [] }] }, { tfOpName: "Conv3D", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }, { tfName: "data_format", name: "dataFormat", type: "string", defaultValue: "NHWC" }, { tfName: "dilations", name: "dilations", type: "number[]" }] }, { tfOpName: "Dilation2D", category: "convolution", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "filter", type: "tensor" }], attrs: [{ tfName: "strides", name: "strides", type: "number[]" }, { tfName: "rates", name: "dilations", type: "number[]" }, { tfName: "padding", name: "pad", type: "string" }] }];
var d0 = {};
Ee(d0, { json: () => QW });
var QW = [{ tfOpName: "Fill", category: "creation", inputs: [{ start: 0, name: "shape", type: "number[]" }, { start: 1, name: "value", type: "number" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype" }] }, { tfOpName: "LinSpace", category: "creation", inputs: [{ start: 0, name: "start", type: "number" }, { start: 1, name: "stop", type: "number" }, { start: 2, name: "num", type: "number" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "OneHot", category: "creation", inputs: [{ start: 0, name: "indices", type: "tensor" }, { start: 1, name: "depth", type: "number" }, { start: 2, name: "onValue", type: "number", defaultValue: 1 }, { start: 3, name: "offValue", type: "number", defaultValue: 0 }], attrs: [{ tfName: "axis", name: "axis", type: "number", notSupported: true }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Ones", category: "creation", inputs: [{ start: 0, name: "shape", type: "number[]" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype" }] }, { tfOpName: "OnesLike", category: "creation", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "dtype", name: "dtype", type: "dtype" }] }, { tfOpName: "RandomUniform", category: "creation", inputs: [{ start: 0, name: "shape", type: "number[]" }], attrs: [{ tfName: "minval", name: "minval", type: "number", defaultValue: 0 }, { tfName: "maxval", name: "maxval", type: "number", defaultValue: 1 }, { tfName: "dtype", name: "dtype", type: "dtype" }, { tfName: "seed", name: "seed", type: "number", defaultValue: 0 }, { tfName: "seed2", name: "seed2", type: "number", defaultValue: 0, notSupported: true }, { tfName: "T", name: "T", type: "number", notSupported: true }] }, { tfOpName: "Range", category: "creation", inputs: [{ start: 0, name: "start", type: "number" }, { start: 1, name: "stop", type: "number" }, { start: 2, name: "step", type: "number", defaultValue: 0 }], attrs: [{ tfName: "Tidx", name: "dtype", type: "dtype" }] }, { tfOpName: "TruncatedNormal", category: "creation", inputs: [{ start: 0, name: "shape", type: "number[]" }], attrs: [{ tfName: "means", name: "mean", type: "number", defaultValue: 0 }, { tfName: "stddev", name: "stdDev", type: "number", defaultValue: 1 }, { tfName: "seed", name: "seed", type: "number" }, { tfName: "seed2", name: "seed2", type: "number", defaultValue: 0, notSupported: true }, { tfName: "dtype", name: "dtype", type: "dtype" }, { tfName: "T", name: "T", type: "number", notSupported: true }] }, { tfOpName: "Zeros", category: "creation", inputs: [{ start: 0, name: "shape", type: "number[]" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype" }] }, { tfOpName: "ZerosLike", category: "creation", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype" }] }, { tfOpName: "Multinomial", category: "creation", inputs: [{ start: 0, name: "logits", type: "tensor" }, { start: 1, name: "numSamples", type: "number" }], attrs: [{ tfName: "seed", name: "seed", type: "number" }, { tfName: "seed2", name: "seed2", type: "number" }, { tfName: "T", name: "dtype", type: "dtype" }, { tfName: "output_dtype", name: "output_dtype", type: "dtype" }] }];
var p0 = {};
Ee(p0, { json: () => ZW });
var ZW = [{ tfOpName: "NonMaxSuppressionV2", category: "dynamic", inputs: [{ start: 0, name: "boxes", type: "tensor" }, { start: 1, name: "scores", type: "tensor" }, { start: 2, name: "maxOutputSize", type: "number" }, { start: 3, name: "iouThreshold", type: "number" }] }, { tfOpName: "NonMaxSuppressionV3", category: "dynamic", inputs: [{ start: 0, name: "boxes", type: "tensor" }, { start: 1, name: "scores", type: "tensor" }, { start: 2, name: "maxOutputSize", type: "number" }, { start: 3, name: "iouThreshold", type: "number" }, { start: 4, name: "scoreThreshold", type: "number" }] }, { tfOpName: "NonMaxSuppressionV4", category: "dynamic", inputs: [{ start: 0, name: "boxes", type: "tensor" }, { start: 1, name: "scores", type: "tensor" }, { start: 2, name: "maxOutputSize", type: "number" }, { start: 3, name: "iouThreshold", type: "number" }, { start: 4, name: "scoreThreshold", type: "number" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }, { tfName: "T_threshold", name: "threshold", type: "dtype", notSupported: true }, { tfName: "pad_to_max_output_size", name: "padToMaxOutputSize", type: "bool" }] }, { tfOpName: "NonMaxSuppressionV5", category: "dynamic", inputs: [{ start: 0, name: "boxes", type: "tensor" }, { start: 1, name: "scores", type: "tensor" }, { start: 2, name: "maxOutputSize", type: "number" }, { start: 3, name: "iouThreshold", type: "number" }, { start: 4, name: "scoreThreshold", type: "number" }, { start: 5, name: "softNmsSigma", type: "number" }] }, { tfOpName: "Where", category: "dynamic", inputs: [{ start: 0, name: "condition", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "ListDiff", category: "dynamic", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "y", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }];
var h0 = {};
Ee(h0, { json: () => JW });
var JW = [{ tfOpName: "LowerBound", category: "evaluation", inputs: [{ start: 0, name: "sortedSequence", type: "tensor" }, { start: 1, name: "values", type: "tensor" }] }, { tfOpName: "TopKV2", category: "evaluation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "k", type: "number" }], attrs: [{ tfName: "sorted", name: "sorted", type: "bool" }] }, { tfOpName: "UpperBound", category: "evaluation", inputs: [{ start: 0, name: "sortedSequence", type: "tensor" }, { start: 1, name: "values", type: "tensor" }] }, { tfOpName: "Unique", category: "evaluation", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "UniqueV2", category: "evaluation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number" }] }];
var f0 = {};
Ee(f0, { json: () => e4 });
var e4 = [{ tfOpName: "PlaceholderWithDefault", category: "graph", inputs: [{ start: 0, name: "default", type: "tensor" }], attrs: [{ tfName: "shape", name: "shape", type: "shape" }, { tfName: "dtype", name: "dtype", type: "dtype" }] }, { tfOpName: "Placeholder", category: "graph", attrs: [{ tfName: "shape", name: "shape", type: "shape" }, { tfName: "dtype", name: "dtype", type: "dtype" }] }, { tfOpName: "Const", category: "graph" }, { tfOpName: "Identity", category: "graph", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "IdentityN", category: "graph", inputs: [{ start: 0, end: 0, name: "x", type: "tensors" }] }, { tfOpName: "Snapshot", category: "graph", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "Rank", category: "graph", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "Size", category: "graph", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "Shape", category: "graph", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "ShapeN", category: "graph", inputs: [{ start: 0, end: 0, name: "x", type: "tensors" }] }, { tfOpName: "Print", category: "graph", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "data", type: "tensors" }], attrs: [{ tfName: "message", name: "message", type: "string" }, { tfName: "first_n", name: "firstN", type: "number", notSupported: true }, { tfName: "summarize", name: "summarize", type: "number", defaultValue: 3 }] }, { tfOpName: "NoOp", category: "graph", inputs: [] }, { tfOpName: "StopGradient", category: "graph", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "FakeQuantWithMinMaxVars", category: "graph", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "min", name: "min", type: "number" }, { tfName: "max", name: "max", type: "number" }] }];
var m0 = {};
Ee(m0, { json: () => t4 });
var t4 = [{ tfOpName: "HashTable", category: "hash_table", inputs: [], attrs: [{ tfName: "shared_name", name: "sharedName", type: "string" }, { tfName: "use_node_name_sharing", name: "useNodeNameSharing", type: "bool" }, { tfName: "key_dtype", name: "keyDType", type: "dtype" }, { tfName: "value_dtype", name: "valueDType", type: "dtype" }] }, { tfOpName: "HashTableV2", category: "hash_table", inputs: [], attrs: [{ tfName: "shared_name", name: "sharedName", type: "string" }, { tfName: "use_node_name_sharing", name: "useNodeNameSharing", type: "bool" }, { tfName: "key_dtype", name: "keyDType", type: "dtype" }, { tfName: "value_dtype", name: "valueDType", type: "dtype" }] }, { tfOpName: "LookupTableImport", category: "hash_table", inputs: [{ start: 0, name: "tableHandle", type: "tensor" }, { start: 1, name: "keys", type: "tensor" }, { start: 2, name: "values", type: "tensor" }], attrs: [{ tfName: "Tin", name: "tIn", type: "dtype", notSupported: true }, { tfName: "Tout", name: "tOut", type: "dtype", notSupported: true }] }, { tfOpName: "LookupTableImportV2", category: "hash_table", inputs: [{ start: 0, name: "tableHandle", type: "tensor" }, { start: 1, name: "keys", type: "tensor" }, { start: 2, name: "values", type: "tensor" }], attrs: [{ tfName: "Tin", name: "tIn", type: "dtype", notSupported: true }, { tfName: "Tout", name: "tOut", type: "dtype", notSupported: true }] }, { tfOpName: "LookupTableFind", category: "hash_table", inputs: [{ start: 0, name: "tableHandle", type: "tensor" }, { start: 1, name: "keys", type: "tensor" }, { start: 2, name: "defaultValue", type: "tensor" }], attrs: [{ tfName: "Tin", name: "tIn", type: "dtype", notSupported: true }, { tfName: "Tout", name: "tOut", type: "dtype", notSupported: true }] }, { tfOpName: "LookupTableFindV2", category: "hash_table", inputs: [{ start: 0, name: "tableHandle", type: "tensor" }, { start: 1, name: "keys", type: "tensor" }, { start: 2, name: "defaultValue", type: "tensor" }], attrs: [{ tfName: "Tin", name: "tIn", type: "dtype", notSupported: true }, { tfName: "Tout", name: "tOut", type: "dtype", notSupported: true }] }, { tfOpName: "LookupTableSize", category: "hash_table", inputs: [{ start: 0, name: "tableHandle", type: "tensor" }] }, { tfOpName: "LookupTableSizeV2", category: "hash_table", inputs: [{ start: 0, name: "tableHandle", type: "tensor" }] }];
var g0 = {};
Ee(g0, { json: () => n4 });
var n4 = [{ tfOpName: "ResizeBilinear", category: "image", inputs: [{ start: 0, name: "images", type: "tensor" }, { start: 1, name: "size", type: "number[]" }], attrs: [{ tfName: "align_corners", name: "alignCorners", type: "bool" }, { tfName: "half_pixel_centers", name: "halfPixelCenters", type: "bool" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "ResizeNearestNeighbor", category: "image", inputs: [{ start: 0, name: "images", type: "tensor" }, { start: 1, name: "size", type: "number[]" }], attrs: [{ tfName: "align_corners", name: "alignCorners", type: "bool" }, { tfName: "half_pixel_centers", name: "halfPixelCenters", type: "bool" }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "CropAndResize", category: "image", inputs: [{ start: 0, name: "image", type: "tensor" }, { start: 1, name: "boxes", type: "tensor" }, { start: 2, name: "boxInd", type: "tensor" }, { start: 3, name: "cropSize", type: "number[]" }], attrs: [{ tfName: "method", name: "method", type: "string" }, { tfName: "extrapolation_value", name: "extrapolationValue", type: "number" }] }, { tfOpName: "ImageProjectiveTransformV3", category: "image", inputs: [{ start: 0, name: "images", type: "tensor" }, { start: 1, name: "transforms", type: "tensor" }, { start: 2, name: "outputShape", type: "number[]" }, { start: 3, name: "fillValue", type: "number" }], attrs: [{ tfName: "interpolation", name: "interpolation", type: "string" }, { tfName: "fill_mode", name: "fillMode", type: "string" }] }];
var b0 = {};
Ee(b0, { json: () => s4 });
var s4 = [{ tfOpName: "Equal", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "NotEqual", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Greater", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "GreaterEqual", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Less", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "LessEqual", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "LogicalAnd", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "LogicalNot", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "LogicalOr", category: "logical", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Select", category: "logical", inputs: [{ start: 0, name: "condition", type: "tensor" }, { start: 1, name: "a", type: "tensor" }, { start: 2, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "SelectV2", category: "logical", inputs: [{ start: 0, name: "condition", type: "tensor" }, { start: 1, name: "a", type: "tensor" }, { start: 2, name: "b", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }];
var y0 = {};
Ee(y0, { json: () => r4 });
var r4 = [{ tfOpName: "_FusedMatMul", category: "matrices", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }, { start: 2, end: 0, name: "args", type: "tensors" }], attrs: [{ tfName: "num_args", name: "numArgs", type: "number" }, { tfName: "fused_ops", name: "fusedOps", type: "string[]", defaultValue: [] }, { tfName: "epsilon", name: "epsilon", type: "number", defaultValue: 1e-4 }, { tfName: "transpose_a", name: "transposeA", type: "bool", defaultValue: false }, { tfName: "transpose_b", name: "transposeB", type: "bool", defaultValue: false }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "MatMul", category: "matrices", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "transpose_a", name: "transposeA", type: "bool", defaultValue: false }, { tfName: "transpose_b", name: "transposeB", type: "bool", defaultValue: false }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "BatchMatMul", category: "matrices", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "adj_x", name: "transposeA", type: "bool", defaultValue: false }, { tfName: "adj_y", name: "transposeB", type: "bool", defaultValue: false }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "BatchMatMulV2", category: "matrices", inputs: [{ start: 0, name: "a", type: "tensor" }, { start: 1, name: "b", type: "tensor" }], attrs: [{ tfName: "adj_x", name: "transposeA", type: "bool", defaultValue: false }, { tfName: "adj_y", name: "transposeB", type: "bool", defaultValue: false }, { tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Transpose", category: "matrices", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "perm", type: "number[]" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "Einsum", category: "matrices", inputs: [{ start: 0, end: 0, name: "tensors", type: "tensors" }], attrs: [{ tfName: "equation", name: "equation", type: "string" }, { tfName: "N", name: "n", type: "number", defaultValue: 2 }, { tfName: "T", name: "dtype", type: "dtype" }] }];
var v0 = {};
Ee(v0, { json: () => a4 });
var a4 = [{ tfOpName: "EuclideanNorm", category: "normalization", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool", defaultValue: false }] }, { tfOpName: "FusedBatchNorm", category: "normalization", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "scale", type: "tensor" }, { start: 2, name: "offset", type: "tensor" }, { start: 3, name: "mean", type: "tensor" }, { start: 4, name: "variance", type: "tensor" }], attrs: [{ tfName: "epsilon", name: "epsilon", type: "number", defaultValue: 1e-3 }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }] }, { tfOpName: "FusedBatchNormV2", category: "normalization", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "scale", type: "tensor" }, { start: 2, name: "offset", type: "tensor" }, { start: 3, name: "mean", type: "tensor" }, { start: 4, name: "variance", type: "tensor" }], attrs: [{ tfName: "epsilon", name: "epsilon", type: "number", defaultValue: 1e-3 }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }] }, { tfOpName: "FusedBatchNormV3", category: "normalization", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "scale", type: "tensor" }, { start: 2, name: "offset", type: "tensor" }, { start: 3, name: "mean", type: "tensor" }, { start: 4, name: "variance", type: "tensor" }], attrs: [{ tfName: "epsilon", name: "epsilon", type: "number", defaultValue: 1e-3 }, { tfName: "data_format", name: "dataFormat", type: "string", notSupported: true }] }, { tfOpName: "LRN", category: "normalization", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "depth_radius", name: "radius", type: "number", defaultValue: 5 }, { tfName: "bias", name: "bias", type: "number", defaultValue: 1 }, { tfName: "alpha", name: "alpha", type: "number", defaultValue: 1 }, { tfName: "beta", name: "beta", type: "number", defaultValue: 0.5 }] }, { tfOpName: "Softmax", category: "normalization", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "LogSoftmax", category: "normalization", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "SparseToDense", category: "normalization", inputs: [{ start: 0, name: "sparseIndices", type: "tensor" }, { start: 1, name: "outputShape", type: "number[]" }, { start: 2, name: "sparseValues", type: "tensor" }, { start: 3, name: "defaultValue", type: "tensor" }], attrs: [{ tfName: "validate_indices", name: "validateIndices", type: "bool", defaultValue: true, notSupported: true }] }];
var x0 = {};
Ee(x0, { json: () => i4 });
var i4 = [{ tfOpName: "Bincount", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "size", type: "number" }, { start: 2, name: "weights", type: "tensor" }] }, { tfOpName: "DenseBincount", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "size", type: "number" }, { start: 2, name: "weights", type: "tensor" }], attrs: [{ tfName: "binary_output", name: "binaryOutput", type: "bool" }] }, { tfOpName: "Max", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool" }] }, { tfOpName: "Mean", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool" }] }, { tfOpName: "Min", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool" }] }, { tfOpName: "Sum", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool" }] }, { tfOpName: "All", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool" }] }, { tfOpName: "Any", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool" }] }, { tfOpName: "ArgMax", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number" }] }, { tfOpName: "ArgMin", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number" }] }, { tfOpName: "Prod", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }], attrs: [{ tfName: "keep_dims", name: "keepDims", type: "bool" }] }, { tfOpName: "Cumprod", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number" }], attrs: [{ tfName: "exclusive", name: "exclusive", type: "bool" }, { tfName: "reverse", name: "reverse", type: "bool" }] }, { tfOpName: "Cumsum", category: "reduction", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number" }], attrs: [{ tfName: "exclusive", name: "exclusive", type: "bool" }, { tfName: "reverse", name: "reverse", type: "bool" }] }];
var w0 = {};
Ee(w0, { json: () => o4 });
var o4 = [{ tfOpName: "ConcatV2", category: "slice_join", inputs: [{ start: 0, end: -1, name: "tensors", type: "tensors" }, { start: -1, name: "axis", type: "number" }], attrs: [{ tfName: "N", name: "n", type: "number", defaultValue: 2 }] }, { tfOpName: "Concat", category: "slice_join", inputs: [{ start: 1, end: 0, name: "tensors", type: "tensors" }, { start: 0, name: "axis", type: "number" }], attrs: [{ tfName: "N", name: "n", type: "number", defaultValue: 2 }] }, { tfOpName: "GatherV2", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "indices", type: "tensor" }, { start: 2, name: "axis", type: "number", defaultValue: 0 }], attrs: [{ tfName: "batch_dims", name: "batchDims", type: "number", defaultValue: 0 }] }, { tfOpName: "Gather", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "indices", type: "tensor" }], attrs: [{ tfName: "validate_indices", name: "validateIndices", type: "bool", notSupported: true }] }, { tfOpName: "Reverse", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "dims", type: "bool[]" }] }, { tfOpName: "ReverseV2", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number[]" }] }, { tfOpName: "Slice", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "begin", type: "number[]" }, { start: 2, name: "size", type: "number[]" }] }, { tfOpName: "StridedSlice", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "begin", type: "number[]" }, { start: 2, name: "end", type: "number[]" }, { start: 3, name: "strides", type: "number[]" }], attrs: [{ tfName: "begin_mask", name: "beginMask", type: "number", defaultValue: 0 }, { tfName: "end_mask", name: "endMask", type: "number", defaultValue: 0 }, { tfName: "new_axis_mask", name: "newAxisMask", type: "number", defaultValue: 0 }, { tfName: "ellipsis_mask", name: "ellipsisMask", type: "number", defaultValue: 0 }, { tfName: "shrink_axis_mask", name: "shrinkAxisMask", type: "number", defaultValue: 0 }] }, { tfOpName: "Pack", category: "slice_join", inputs: [{ start: 0, end: 0, name: "tensors", type: "tensors" }], attrs: [{ tfName: "axis", name: "axis", type: "number", defaultValue: 0 }] }, { tfOpName: "Unpack", category: "slice_join", inputs: [{ start: 0, name: "tensor", type: "tensor" }], attrs: [{ tfName: "axis", name: "axis", type: "number", defaultValue: 0 }, { tfName: "num", name: "num", type: "number", defaultValue: 0, notSupported: true }] }, { tfOpName: "Tile", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "reps", type: "number[]" }] }, { tfOpName: "Split", category: "slice_join", inputs: [{ start: 0, name: "axis", type: "number", defaultValue: 0 }, { start: 1, name: "x", type: "tensor" }], attrs: [{ tfName: "num_split", name: "numOrSizeSplits", type: "number", defaultValue: 1 }] }, { tfOpName: "SplitV", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "numOrSizeSplits", type: "number[]" }, { start: 2, name: "axis", type: "number", defaultValue: 0 }] }, { tfOpName: "ScatterNd", category: "slice_join", inputs: [{ start: 0, name: "indices", type: "tensor" }, { start: 1, name: "values", type: "tensor" }, { start: 2, name: "shape", type: "number[]" }] }, { tfOpName: "GatherNd", category: "slice_join", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "indices", type: "tensor" }] }, { tfOpName: "SparseToDense", category: "slice_join", inputs: [{ start: 0, name: "sparseIndices", type: "tensor" }, { start: 1, name: "outputShape", type: "number[]" }, { start: 2, name: "sparseValues", type: "tensor" }, { start: 3, name: "defaultValue", type: "tensor" }], attrs: [{ tfName: "validate_indices", name: "validateIndices", type: "bool", defaultValue: false, notSupported: true }] }];
var k0 = {};
Ee(k0, { json: () => u4 });
var u4 = [{ tfOpName: "SparseFillEmptyRows", category: "sparse", inputs: [{ start: 0, name: "indices", type: "tensor" }, { start: 1, name: "values", type: "tensor" }, { start: 2, name: "denseShape", type: "tensor" }, { start: 3, name: "defaultValue", type: "tensor" }] }, { tfOpName: "SparseReshape", category: "sparse", inputs: [{ start: 0, name: "inputIndices", type: "tensor" }, { start: 1, name: "inputShape", type: "tensor" }, { start: 2, name: "newShape", type: "tensor" }], attrs: [{ tfName: "T", name: "dtype", type: "dtype", notSupported: true }] }, { tfOpName: "SparseSegmentMean", category: "sparse", inputs: [{ start: 0, name: "data", type: "tensor" }, { start: 1, name: "indices", type: "tensor" }, { start: 2, name: "segmentIds", type: "tensor" }] }, { tfOpName: "SparseSegmentSum", category: "sparse", inputs: [{ start: 0, name: "data", type: "tensor" }, { start: 1, name: "indices", type: "tensor" }, { start: 2, name: "segmentIds", type: "tensor" }] }];
var S0 = {};
Ee(S0, { json: () => l4 });
var l4 = [{ tfOpName: "FFT", category: "spectral", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "IFFT", category: "spectral", inputs: [{ start: 0, name: "x", type: "tensor" }] }, { tfOpName: "RFFT", category: "spectral", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "fft_length", type: "number", notSupported: true }] }, { tfOpName: "IRFFT", category: "spectral", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "fft_length", type: "number", notSupported: true }] }];
var I0 = {};
Ee(I0, { json: () => c4 });
var c4 = [{ tfOpName: "StringNGrams", category: "string", inputs: [{ start: 0, name: "data", type: "tensor" }, { start: 1, name: "dataSplits", type: "tensor" }], attrs: [{ tfName: "separator", name: "separator", type: "string" }, { tfName: "ngram_widths", name: "nGramWidths", type: "number[]" }, { tfName: "left_pad", name: "leftPad", type: "string" }, { tfName: "right_pad", name: "rightPad", type: "string" }, { tfName: "pad_width", name: "padWidth", type: "number" }, { tfName: "preserve_short_sequences", name: "preserveShortSequences", type: "bool" }], outputs: ["ngrams", "ngrams_splits"] }, { tfOpName: "StringSplit", category: "string", inputs: [{ start: 0, name: "input", type: "tensor" }, { start: 1, name: "delimiter", type: "tensor" }], attrs: [{ tfName: "skip_empty", name: "skipEmpty", type: "bool" }], outputs: ["indices", "values", "shape"] }, { tfOpName: "StringToHashBucketFast", category: "string", inputs: [{ start: 0, name: "input", type: "tensor" }], attrs: [{ tfName: "num_buckets", name: "numBuckets", type: "number" }] }];
var C0 = {};
Ee(C0, { json: () => d4 });
var d4 = [{ tfOpName: "Cast", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "SrcT", name: "sdtype", type: "dtype", notSupported: true }, { tfName: "DstT", name: "dtype", type: "dtype" }] }, { tfOpName: "ExpandDims", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "axis", type: "number" }] }, { tfOpName: "MirrorPad", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "padding", type: "number[]" }], attrs: [{ tfName: "mode", name: "mode", type: "string" }] }, { tfOpName: "Pad", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "padding", type: "number[]" }], attrs: [{ tfName: "constant_value", name: "constantValue", type: "number", defaultValue: 0 }] }, { tfOpName: "PadV2", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "padding", type: "number[]" }, { start: 2, name: "constantValue", type: "number", defaultValue: 0 }] }, { tfOpName: "Reshape", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "shape", type: "number[]" }] }, { tfOpName: "Squeeze", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "axis", tfDeprecatedName: "squeeze_dims", name: "axis", type: "number[]" }] }, { tfOpName: "SpaceToBatchND", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "blockShape", type: "number[]" }, { start: 2, name: "paddings", type: "number[]" }] }, { tfOpName: "BatchToSpaceND", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "blockShape", type: "number[]" }, { start: 2, name: "crops", type: "number[]" }] }, { tfOpName: "DepthToSpace", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }], attrs: [{ tfName: "block_size", name: "blockSize", type: "number" }, { tfName: "data_format", name: "dataFormat", type: "string" }] }, { tfOpName: "BroadcastTo", category: "transformation", inputs: [{ start: 0, name: "x", type: "tensor" }, { start: 1, name: "shape", type: "number[]" }], attrs: [] }, { tfOpName: "BroadcastArgs", category: "transformation", inputs: [{ start: 0, name: "s0", type: "tensor" }, { start: 1, name: "s1", type: "tensor" }], attrs: [] }];
var Zx = class {
  static get Instance() {
    return this._instance || (this._instance = new this());
  }
  constructor() {
    let e = [o0, u0, l0, c0, d0, p0, h0, f0, m0, g0, b0, y0, v0, x0, w0, k0, S0, I0, C0], t = [].concat(...e.map((n) => n.json));
    this.opMappers = t.reduce((n, s) => (n[s.tfOpName] = s, n), {});
  }
  transformGraph(e, t = {}) {
    let n = e.node, s = [], r = [], a = [], i = n.reduce((f, m) => (f[m.name] = this.mapNode(m), m.op.startsWith("Placeholder") ? s.push(f[m.name]) : m.op === "Const" ? r.push(f[m.name]) : (m.input == null || m.input.length === 0) && a.push(f[m.name]), f), {}), o = [], u = [], l = {}, c = {};
    t != null && (l = this.mapSignatureEntries(t.inputs), c = this.mapSignatureEntries(t.outputs));
    let p = Object.keys(i);
    p.forEach((f) => {
      let m = i[f];
      m.inputNames.forEach((g, b) => {
        let [y, , v] = Ts(g), x = i[y];
        if (x.outputs != null) {
          let k = x.outputs.indexOf(v);
          if (k !== -1) {
            let I = `${y}:${k}`;
            m.inputNames[b] = I;
          }
        }
        m.inputs.push(x), x.children.push(m);
      });
    }), Object.keys(c).length === 0 ? p.forEach((f) => {
      let m = i[f];
      m.children.length === 0 && u.push(m);
    }) : Object.keys(c).forEach((f) => {
      let [m] = Ts(f), g = i[m];
      g != null && (g.signatureKey = c[f], u.push(g));
    }), Object.keys(l).length > 0 ? Object.keys(l).forEach((f) => {
      let [m] = Ts(f), g = i[m];
      g && (g.signatureKey = l[f], o.push(g));
    }) : o = s;
    let d = {};
    e.library != null && e.library.function != null && (d = e.library.function.reduce((f, m) => (f[m.signature.name] = this.mapFunction(m), f), {}));
    let h = { nodes: i, inputs: o, outputs: u, weights: r, placeholders: s, signature: t, functions: d };
    return a.length > 0 && (h.initNodes = a), h;
  }
  mapSignatureEntries(e) {
    return Object.keys(e || {}).reduce((t, n) => (t[e[n].name] = n, t), {});
  }
  mapNode(e) {
    let t = i0(e.op) || this.opMappers[e.op] || {};
    e.attr == null && (e.attr = {});
    let n = { name: e.name, op: e.op, category: t.category, inputNames: (e.input || []).map((s) => s.startsWith("^") ? s.slice(1) : s), inputs: [], children: [], inputParams: {}, attrParams: {}, rawAttrs: e.attr, outputs: t.outputs };
    return t.inputs != null && (n.inputParams = t.inputs.reduce((s, r) => (s[r.name] = { type: r.type, inputIndexStart: r.start, inputIndexEnd: r.end }, s), {})), t.attrs != null && (n.attrParams = t.attrs.reduce((s, r) => {
      let a = r.type, i;
      switch (r.type) {
        case "string":
          i = Dm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Dm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "string[]":
          i = Bm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Bm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "number":
          i = Om(e.attr, r.tfName, r.defaultValue || 0), i === void 0 && !!r.tfDeprecatedName && (i = Om(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "number[]":
          i = Lm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Lm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "bool":
          i = Fm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Fm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "bool[]":
          i = Wm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Wm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "shape":
          i = Mm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Mm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "shape[]":
          i = Vm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Vm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "dtype":
          i = Pm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Pm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "dtype[]":
          i = zm(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = zm(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "func":
          i = Jx(e.attr, r.tfName, r.defaultValue), i === void 0 && !!r.tfDeprecatedName && (i = Jx(e.attr, r.tfDeprecatedName, r.defaultValue));
          break;
        case "tensor":
        case "tensors":
          break;
        default:
          throw new Error(`Unsupported param type: ${r.type} for op: ${e.op}`);
      }
      return s[r.name] = { value: i, type: a }, s;
    }, {})), n;
  }
  mapFunction(e) {
    let t = e.nodeDef, n = [], s = [], r = {};
    t != null && (r = t.reduce((c, p) => (c[p.name] = this.mapNode(p), p.op === "Const" && s.push(c[p.name]), c), {}));
    let a = [], i = [];
    e.signature.inputArg.forEach((c) => {
      let [p] = Ts(c.name), d = { name: p, op: "Placeholder", inputs: [], inputNames: [], category: "graph", inputParams: {}, attrParams: { dtype: { value: sv(c.type), type: "dtype" } }, children: [] };
      d.signatureKey = c.name, a.push(d), r[p] = d;
    }), Object.keys(r).forEach((c) => {
      let p = r[c];
      p.inputNames.forEach((d, h) => {
        let [f, , m] = Ts(d), g = r[f];
        if (g.outputs != null) {
          let b = g.outputs.indexOf(m);
          if (b !== -1) {
            let y = `${f}:${b}`;
            p.inputNames[h] = y;
          }
        }
        p.inputs.push(g), g.children.push(p);
      });
    });
    let u = e.ret;
    e.signature.outputArg.forEach((c) => {
      let [p, d] = Ts(u[c.name]), h = r[p];
      h != null && (h.defaultOutput = d, i.push(h));
    });
    let l = this.mapArgsToSignature(e);
    return { nodes: r, inputs: a, outputs: i, weights: s, placeholders: n, signature: l };
  }
  mapArgsToSignature(e) {
    return { methodName: e.signature.name, inputs: e.signature.inputArg.reduce((t, n) => (t[n.name] = this.mapArgToTensorInfo(n), t), {}), outputs: e.signature.outputArg.reduce((t, n) => (t[n.name] = this.mapArgToTensorInfo(n, e.ret), t), {}) };
  }
  mapArgToTensorInfo(e, t) {
    let n = e.name;
    return t != null && (n = t[n]), { name: n, dtype: e.type };
  }
};
function p4(e) {
  let t = K().global;
  if (typeof t.atob != "undefined")
    return t.atob(e);
  if (typeof Buffer != "undefined")
    return new Buffer(e, "base64").toString();
  throw new Error("Unable to decode base64 in this environment. Missing built-in atob() or Buffer()");
}
function N0(e, t) {
  let n = Array.isArray(e) ? String.fromCharCode.apply(null, e) : p4(e);
  return t ? n : n.toLowerCase();
}
function Dm(e, t, n, s = false) {
  let r = e[t];
  return r != null ? N0(r.s, s) : n;
}
function Fm(e, t, n) {
  let s = e[t];
  return s ? s.b : n;
}
function Om(e, t, n) {
  let s = e[t] || {}, r = s.i != null ? s.i : s.f != null ? s.f : n;
  return typeof r == "number" ? r : parseInt(r, 10);
}
function sv(e) {
  switch (typeof e == "string" && (e = a0[e]), e) {
    case 1:
    case 19:
      return "float32";
    case 3:
    case 9:
    case 6:
    case 4:
      return "int32";
    case 10:
      return "bool";
    case 2:
      return "float32";
    case 7:
      return "string";
    default:
      return null;
  }
}
function Jx(e, t, n) {
  let s = e[t];
  return s && s.func ? s.func.name : n;
}
function Pm(e, t, n) {
  let s = e[t];
  return s && s.type ? sv(s.type) : n;
}
function zm(e, t, n) {
  let s = e[t];
  return s && s.list && s.list.type ? s.list.type.map((r) => sv(r)) : n;
}
function T0(e) {
  if (!e.unknownRank)
    return e.dim != null ? e.dim.map((t) => typeof t.size == "number" ? t.size : parseInt(t.size, 10)) : [];
}
function Mm(e, t, n) {
  let s = e[t];
  return s && s.shape ? T0(s.shape) : n;
}
function Lm(e, t, n) {
  let s = e[t];
  return s ? ((s.list.f && s.list.f.length ? s.list.f : s.list.i) || []).map((r) => typeof r == "number" ? r : parseInt(r, 10)) : n;
}
function Bm(e, t, n, s = false) {
  let r = e[t];
  return r && r.list && r.list.s ? r.list.s.map((a) => N0(a, s)) : n;
}
function Vm(e, t, n) {
  let s = e[t];
  return s && s.list && s.list.shape ? s.list.shape.map((r) => T0(r)) : n;
}
function Wm(e, t, n) {
  let s = e[t];
  return s && s.list && s.list.b ? s.list.b : n;
}
var h4 = class {
  constructor(e, t, n) {
    this.node = e, this.tensorMap = t, this.context = n, this.inputs = [], this.attrs = {}, this.inputs = e.inputNames.map((s) => this.getInput(s)), e.rawAttrs != null && (this.attrs = Object.keys(e.rawAttrs).reduce((s, r) => (s[r] = this.getAttr(r), s), {}));
  }
  getInput(e) {
    return un(e, this.tensorMap, this.context);
  }
  getAttr(e, t) {
    let n = this.node.rawAttrs[e];
    if (n.tensor != null)
      return un(e, this.tensorMap, this.context);
    if (n.i != null || n.f != null)
      return Om(this.node.rawAttrs, e, t);
    if (n.s != null)
      return Dm(this.node.rawAttrs, e, t);
    if (n.b != null)
      return Fm(this.node.rawAttrs, e, t);
    if (n.shape != null)
      return Mm(this.node.rawAttrs, e, t);
    if (n.type != null)
      return Pm(this.node.rawAttrs, e, t);
    if (n.list != null) {
      if (n.list.i != null || n.list.f != null)
        return Lm(this.node.rawAttrs, e, t);
      if (n.list.s != null)
        return Bm(this.node.rawAttrs, e, t);
      if (n.list.shape != null)
        return Vm(this.node.rawAttrs, e, t);
      if (n.list.b != null)
        return Wm(this.node.rawAttrs, e, t);
      if (n.list.type != null)
        return zm(this.node.rawAttrs, e, t);
    }
    return t;
  }
};
var f4 = (e, t, n) => {
  switch (e.op) {
    case "BiasAdd":
    case "AddV2":
    case "Add":
      return [ie(S("a", e, t, n), S("b", e, t, n))];
    case "AddN":
      return [nE(S("tensors", e, t, n))];
    case "FloorMod":
    case "Mod":
      return [zD(S("a", e, t, n), S("b", e, t, n))];
    case "Mul":
      return [V(S("a", e, t, n), S("b", e, t, n))];
    case "RealDiv":
    case "Div":
      return [xe(S("a", e, t, n), S("b", e, t, n))];
    case "DivNoNan":
      return [NR(S("a", e, t, n), S("b", e, t, n))];
    case "FloorDiv":
      return [tS(S("a", e, t, n), S("b", e, t, n))];
    case "Sub":
      return [ge(S("a", e, t, n), S("b", e, t, n))];
    case "Minimum":
      return [Ip(S("a", e, t, n), S("b", e, t, n))];
    case "Maximum":
      return [$r(S("a", e, t, n), S("b", e, t, n))];
    case "Pow":
      return [ha(S("a", e, t, n), S("b", e, t, n))];
    case "SquaredDifference":
      return [DS(S("a", e, t, n), S("b", e, t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var m4 = (e, t, n) => {
  switch (e.op) {
    case "Abs":
    case "ComplexAbs":
      return [Lt(S("x", e, t, n))];
    case "Acos":
      return [ZA(S("x", e, t, n))];
    case "Acosh":
      return [eE(S("x", e, t, n))];
    case "Asin":
      return [lE(S("x", e, t, n))];
    case "Asinh":
      return [dE(S("x", e, t, n))];
    case "Atan":
      return [hE(S("x", e, t, n))];
    case "Atan2":
      return [mE(S("x", e, t, n), S("y", e, t, n))];
    case "Atanh":
      return [bE(S("x", e, t, n))];
    case "Ceil":
      return [qE(S("x", e, t, n))];
    case "Complex":
      return [ua(S("real", e, t, n), S("imag", e, t, n))];
    case "Cos":
      return [eb(S("x", e, t, n))];
    case "Cosh":
      return [pS(S("x", e, t, n))];
    case "Elu":
      return [wp(S("x", e, t, n))];
    case "Erf":
      return [RR(S("x", e, t, n))];
    case "Exp":
      return [Yn(S("x", e, t, n))];
    case "Expm1":
      return [jR(S("x", e, t, n))];
    case "Floor":
      return [kp(S("x", e, t, n))];
    case "Log":
      return [Qn(S("x", e, t, n))];
    case "Log1p":
      return [ib(S("x", e, t, n))];
    case "Imag":
      return [rb(S("x", e, t, n))];
    case "Neg":
      return [kt(S("x", e, t, n))];
    case "Reciprocal":
      return [d3(S("x", e, t, n))];
    case "Real":
      return [Cd(S("x", e, t, n))];
    case "Relu":
      return [Xs(S("x", e, t, n))];
    case "Round":
      return [NS(S("x", e, t, n))];
    case "Selu":
      return [$S(S("x", e, t, n))];
    case "Sigmoid":
      return [Hs(S("x", e, t, n))];
    case "Sin":
      return [_S(S("x", e, t, n))];
    case "Sign":
      return [T3(S("x", e, t, n))];
    case "Sinh":
      return [AS(S("x", e, t, n))];
    case "Softplus":
      return [Bl(S("x", e, t, n))];
    case "Sqrt":
      return [dn(S("x", e, t, n))];
    case "Square":
      return [ct(S("x", e, t, n))];
    case "Tanh":
      return [Yu(S("x", e, t, n))];
    case "Tan":
      return [j3(S("x", e, t, n))];
    case "ClipByValue":
      return [Vn(S("x", e, t, n), S("clipValueMin", e, t, n), S("clipValueMax", e, t, n))];
    case "Relu6":
      return [CS(S("x", e, t, n))];
    case "Rsqrt":
      return [TS(un(e.inputNames[0], t, n))];
    case "Prod":
      return [IS(S("x", e, t, n), S("axes", e, t, n))];
    case "LeakyRelu":
      return [ab(S("x", e, t, n), S("alpha", e, t, n))];
    case "Prelu":
      return [db(S("x", e, t, n), S("alpha", e, t, n))];
    case "IsNan":
      return [rD(un(e.inputNames[0], t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
function qn(e, t, n = "") {
  if (!(typeof e == "number" || typeof t == "number")) {
    w.assert(e.length === t.length, () => n + ` Shapes ${e} and ${t} must match`);
    for (let s = 0; s < e.length; s++) {
      let r = e[s], a = t[s];
      w.assert(r < 0 || a < 0 || r === a, () => n + ` Shapes ${e} and ${t} must match`);
    }
  }
}
function ew(e) {
  return !(typeof e == "number" || e.some((t) => t < 0));
}
function Au(e, t, n) {
  let s = Um(e, n), r = !ew(s);
  if (r && t.length === 0)
    throw new Error(`Tried to calculate elements of an empty list with non-fully-defined elementShape: ${s}`);
  if (r && t.forEach((a) => {
    s = Um(a.shape, s);
  }), !ew(s))
    throw new Error(`Non-fully-defined elementShape: ${s}`);
  return s;
}
function Um(e, t) {
  if (typeof e == "number")
    return t;
  if (typeof t == "number")
    return e;
  if (e.length !== t.length)
    throw new Error(`Incompatible ranks during merge: ${e} vs. ${t}`);
  let n = [];
  for (let s = 0; s < e.length; ++s) {
    let r = e[s], a = t[s];
    if (r >= 0 && a >= 0 && r !== a)
      throw new Error(`Incompatible shape during merge: ${e} vs. ${t}`);
    n[s] = r >= 0 ? r : a;
  }
  return n;
}
var g4 = class {
  constructor(e, t, n, s, r, a, i) {
    this.name = e, this.dtype = t, this.maxSize = n, this.elementShape = s, this.identicalElementShapes = r, this.dynamicSize = a, this.clearAfterRead = i, this.tensors = [], this.closed_ = false, this.idTensor = we(0), qt(this.idTensor);
  }
  get id() {
    return this.idTensor.id;
  }
  get closed() {
    return this.closed_;
  }
  clearAndClose(e) {
    this.tensors.forEach((t) => {
      (e == null || !e.has(t.tensor.id)) && t.tensor.dispose();
    }), this.tensors = [], this.closed_ = true, this.idTensor.dispose();
  }
  size() {
    return this.tensors.length;
  }
  read(e) {
    if (this.closed_)
      throw new Error(`TensorArray ${this.name} has already been closed.`);
    if (e < 0 || e >= this.size())
      throw new Error(`Tried to read from index ${e}, but array size is: ${this.size()}`);
    let t = this.tensors[e];
    if (t.cleared)
      throw new Error(`TensorArray ${this.name}: Could not read index ${e} twice because it was cleared after a previous read (perhaps try setting clear_after_read = false?).`);
    return this.clearAfterRead && (t.cleared = true), t.read = true, t.tensor;
  }
  readMany(e) {
    return e.map((t) => this.read(t));
  }
  write(e, t) {
    if (this.closed_)
      throw new Error(`TensorArray ${this.name} has already been closed.`);
    if (e < 0 || !this.dynamicSize && e >= this.maxSize)
      throw new Error(`Tried to write to index ${e}, but array is not resizeable and size is: ${this.maxSize}`);
    let n = this.tensors[e] || {};
    if (t.dtype !== this.dtype)
      throw new Error(`TensorArray ${this.name}: Could not write to TensorArray index ${e},
          because the value dtype is ${t.dtype}, but TensorArray dtype is ${this.dtype}.`);
    if (this.size() === 0 && (this.elementShape == null || this.elementShape.length === 0) && (this.elementShape = t.shape), qn(this.elementShape, t.shape, `TensorArray ${this.name}: Could not write to TensorArray index ${e}.`), n.read)
      throw new Error(`TensorArray ${this.name}: Could not write to TensorArray index ${e}, because it has already been read.`);
    if (n.written)
      throw new Error(`TensorArray ${this.name}: Could not write to TensorArray index ${e}, because it has already been written.`);
    n.tensor = t, qt(t), n.written = true, this.tensors[e] = n;
  }
  writeMany(e, t) {
    if (e.length !== t.length)
      throw new Error(`TensorArray ${this.name}: could not write multiple tensors,because the index size: ${e.length} is not the same as tensors size: ${t.length}.`);
    e.forEach((n, s) => this.write(n, t[s]));
  }
  gather(e, t) {
    if (!!t && t !== this.dtype)
      throw new Error(`TensorArray dtype is ${this.dtype} but gather requested dtype ${t}`);
    if (e)
      e = e.slice(0, this.size());
    else {
      e = [];
      for (let s = 0; s < this.size(); s++)
        e.push(s);
    }
    if (e.length === 0)
      return ms([], [0].concat(this.elementShape));
    let n = this.readMany(e);
    return qn(this.elementShape, n[0].shape, "TensorArray shape mismatch: "), es(n, 0);
  }
  concat(e) {
    if (!!e && e !== this.dtype)
      throw new Error(`TensorArray dtype is ${this.dtype} but concat requested dtype ${e}`);
    if (this.size() === 0)
      return ms([], [0].concat(this.elementShape));
    let t = [];
    for (let s = 0; s < this.size(); s++)
      t.push(s);
    let n = this.readMany(t);
    return qn(this.elementShape, n[0].shape, `TensorArray shape mismatch: tensor array shape (${this.elementShape}) vs first tensor shape (${n[0].shape})`), Ot(n, 0);
  }
  scatter(e, t) {
    if (t.dtype !== this.dtype)
      throw new Error(`TensorArray dtype is ${this.dtype} but tensor has dtype ${t.dtype}`);
    if (e.length !== t.shape[0])
      throw new Error(`Expected len(indices) == tensor.shape[0], but saw: ${e.length} vs. ${t.shape[0]}`);
    let n = Math.max(...e);
    if (!this.dynamicSize && n >= this.maxSize)
      throw new Error(`Max index must be < array size (${n}  vs. ${this.maxSize})`);
    this.writeMany(e, Fs(t, 0));
  }
  split(e, t) {
    if (t.dtype !== this.dtype)
      throw new Error(`TensorArray dtype is ${this.dtype} but tensor has dtype ${t.dtype}`);
    let n = 0, s = e.map((o) => (n += o, n));
    if (n !== t.shape[0])
      throw new Error(`Expected sum of lengths to be equal to
          tensor.shape[0], but sum of lengths is
        ${n}, and tensor's shape is: ${t.shape}`);
    if (!this.dynamicSize && e.length !== this.maxSize)
      throw new Error(`TensorArray's size is not equal to the size of lengths (${this.maxSize} vs. ${e.length}), and the TensorArray is not marked as dynamically resizeable`);
    let r = n === 0 ? 0 : t.size / n, a = [];
    j(() => {
      t = U(t, [1, n, r]);
      for (let o = 0; o < e.length; ++o) {
        let u = o === 0 ? 0 : s[o - 1], l = [0, u, 0], c = [1, e[o], r];
        a[o] = U(qe(t, l, c), this.elementShape);
      }
      return a;
    });
    let i = [];
    for (let o = 0; o < e.length; o++)
      i[o] = o;
    this.writeMany(i, a);
  }
};
var ro = class {
  constructor(e, t, n, s = -1) {
    this.tensors = e, this.elementShape = t, this.elementDtype = n, e != null && e.forEach((r) => {
      if (n !== r.dtype)
        throw new Error(`Invalid data types; op elements ${n}, but list elements ${r.dtype}`);
      qn(t, r.shape, "TensorList shape mismatch: "), qt(r);
    }), this.idTensor = we(0), this.maxNumElements = s, qt(this.idTensor);
  }
  get id() {
    return this.idTensor.id;
  }
  copy() {
    return new ro([...this.tensors], this.elementShape, this.elementDtype);
  }
  clearAndClose(e) {
    this.tensors.forEach((t) => {
      (e == null || !e.has(t.id)) && t.dispose();
    }), this.tensors.length = 0, this.idTensor.dispose();
  }
  size() {
    return this.tensors.length;
  }
  stack(e, t, n = -1) {
    if (t !== this.elementDtype)
      throw new Error(`Invalid data types; op elements ${t}, but list elements ${this.elementDtype}`);
    if (n !== -1 && this.tensors.length !== n)
      throw new Error(`Operation expected a list with ${n} elements but got a list with ${this.tensors.length} elements.`);
    qn(e, this.elementShape, "TensorList shape mismatch: ");
    let s = Au(this.elementShape, this.tensors, e);
    return j(() => {
      let r = this.tensors.map((a) => U(a, s));
      return es(r, 0);
    });
  }
  popBack(e, t) {
    if (t !== this.elementDtype)
      throw new Error(`Invalid data types; op elements ${t}, but list elements ${this.elementDtype}`);
    if (this.size() === 0)
      throw new Error("Trying to pop from an empty list.");
    let n = Au(this.elementShape, this.tensors, e), s = this.tensors.pop();
    return qn(s.shape, e, "TensorList shape mismatch: "), U(s, n);
  }
  pushBack(e) {
    if (e.dtype !== this.elementDtype)
      throw new Error(`Invalid data types; op elements ${e.dtype}, but list elements ${this.elementDtype}`);
    if (qn(e.shape, this.elementShape, "TensorList shape mismatch: "), this.maxNumElements === this.size())
      throw new Error("Trying to push element into a full list.");
    qt(e), this.tensors.push(e);
  }
  resize(e) {
    if (e < 0)
      throw new Error(`TensorListResize expects size to be non-negative. Got: ${e}`);
    if (this.maxNumElements !== -1 && e > this.maxNumElements)
      throw new Error(`TensorListResize input size ${e} is greater maxNumElement ${this.maxNumElements}.`);
    let t = new ro([], this.elementShape, this.elementDtype, this.maxNumElements);
    t.tensors.length = e;
    for (let n = 0; n < Math.min(this.tensors.length, e); ++n)
      t.tensors[n] = this.tensors[n];
    return t;
  }
  getItem(e, t, n) {
    if (n !== this.elementDtype)
      throw new Error(`Invalid data types; op elements ${n}, but list elements ${this.elementDtype}`);
    if (e < 0 || e > this.tensors.length)
      throw new Error(`Trying to access element ${e} in a list with ${this.tensors.length} elements.`);
    if (this.tensors[e] == null)
      throw new Error(`element at index ${e} is null.`);
    qn(this.tensors[e].shape, t, "TensorList shape mismatch: ");
    let s = Au(this.elementShape, this.tensors, t);
    return U(this.tensors[e], s);
  }
  setItem(e, t) {
    if (t.dtype !== this.elementDtype)
      throw new Error(`Invalid data types; op elements ${t.dtype}, but list elements ${this.elementDtype}`);
    if (e < 0 || this.maxNumElements !== -1 && e >= this.maxNumElements)
      throw new Error(`Trying to set element ${e} in a list with max ${this.maxNumElements} elements.`);
    qn(this.elementShape, t.shape, "TensorList shape mismatch: "), qt(t), this.tensors[e] = t;
  }
  gather(e, t, n) {
    if (t !== this.elementDtype)
      throw new Error(`Invalid data types; op elements ${t}, but list elements ${this.elementDtype}`);
    qn(this.elementShape, n, "TensorList shape mismatch: "), e = e.slice(0, this.size());
    let s = Au(this.elementShape, this.tensors, n);
    return e.length === 0 ? ms([], [0].concat(s)) : j(() => {
      let r = e.map((a) => U(this.tensors[a], s));
      return es(r, 0);
    });
  }
  concat(e, t) {
    if (!!e && e !== this.elementDtype)
      throw new Error(`TensorList dtype is ${this.elementDtype} but concat requested dtype ${e}`);
    qn(this.elementShape, t, "TensorList shape mismatch: ");
    let n = Au(this.elementShape, this.tensors, t);
    return this.size() === 0 ? ms([], [0].concat(n)) : j(() => {
      let s = this.tensors.map((r) => U(r, n));
      return Ot(s, 0);
    });
  }
};
function b4(e, t, n) {
  let s = e.dtype;
  if (e.shape.length < 1)
    throw new Error(`Tensor must be at least a vector, but saw shape: ${e.shape}`);
  if (e.dtype !== n)
    throw new Error(`Invalid data types; op elements ${e.dtype}, but list elements ${n}`);
  let r = e.shape.slice(1);
  qn(r, t, "TensorList shape mismatch: ");
  let a = Fs(e);
  return new ro(a, t, s);
}
function y4(e, t, n) {
  return new ro([], e, t, n);
}
function v4(e, t, n, s) {
  if (t.length !== e.shape[0])
    throw new Error(`Expected len(indices) == tensor.shape[0], but saw: ${t.length} vs. ${e.shape[0]}`);
  let r = Math.max(...t);
  if (s != null && s !== -1 && r >= s)
    throw new Error(`Max index must be < array size (${r}  vs. ${s})`);
  let a = new ro([], n, e.dtype, s), i = Fs(e, 0);
  return t.forEach((o, u) => {
    a.setItem(o, i[u]);
  }), a;
}
function x4(e, t, n) {
  let s = 0, r = t.map((c) => (s += c, s));
  if (s !== e.shape[0])
    throw new Error(`Expected sum of lengths to be equal to
          tensor.shape[0], but sum of lengths is
        ${s}, and tensor's shape is: ${e.shape}`);
  let a = e.shape.slice(1), i = Um(a, n), o = s === 0 ? 0 : e.size / s, u = j(() => {
    let c = [];
    e = U(e, [1, s, o]);
    for (let p = 0; p < t.length; ++p) {
      let d = p === 0 ? 0 : r[p - 1], h = [0, d, 0], f = [1, t[p], o];
      c[p] = U(qe(e, h, f), i);
    }
    return e.dispose(), c;
  }), l = new ro([], n, e.dtype, t.length);
  for (let c = 0; c < u.length; c++)
    l.setItem(c, u[c]);
  return l;
}
var w4 = async (e, t, n) => {
  switch (e.op) {
    case "If":
    case "StatelessIf": {
      let s = S("thenBranch", e, t, n), r = S("elseBranch", e, t, n), a = S("cond", e, t, n), i = S("args", e, t, n);
      return (await a.data())[0] ? n.functionMap[s].executeFunctionAsync(i, n.tensorArrayMap, n.tensorListMap) : n.functionMap[r].executeFunctionAsync(i, n.tensorArrayMap, n.tensorListMap);
    }
    case "While":
    case "StatelessWhile": {
      let s = S("body", e, t, n), r = S("cond", e, t, n), a = S("args", e, t, n), i = await n.functionMap[r].executeFunctionAsync(a, n.tensorArrayMap, n.tensorListMap), o = a.map((c) => c.id), u = await i[0].data();
      i.forEach((c) => {
        !c.kept && o.indexOf(c.id) === -1 && c.dispose();
      });
      let l = a;
      for (; u[0]; ) {
        let c = l;
        l = await n.functionMap[s].executeFunctionAsync(l, n.tensorArrayMap, n.tensorListMap);
        let p = l.map((h) => h.id);
        c.forEach((h) => {
          !h.kept && o.indexOf(h.id) === -1 && p.indexOf(h.id) === -1 && h.dispose();
        });
        let d = await n.functionMap[r].executeFunctionAsync(l, n.tensorArrayMap, n.tensorListMap);
        u = await d[0].data(), d.forEach((h) => {
          !h.kept && o.indexOf(h.id) === -1 && p.indexOf(h.id) === -1 && h.dispose();
        });
      }
      return l;
    }
    case "LoopCond": {
      let s = S("pred", e, t, n);
      return [Ws(s)];
    }
    case "Switch": {
      let s = S("pred", e, t, n), r = S("data", e, t, n);
      return r.kept || (r = Ws(r)), (await s.data())[0] ? [void 0, r] : [r, void 0];
    }
    case "Merge": {
      let s = e.inputNames.find((r) => un(r, t, n) !== void 0);
      if (s) {
        let r = un(s, t, n);
        return [Ws(r)];
      }
      return;
    }
    case "Enter": {
      let s = S("frameName", e, t, n), r = S("tensor", e, t, n);
      return n.enterFrame(s), [Ws(r)];
    }
    case "Exit": {
      let s = S("tensor", e, t, n);
      return n.exitFrame(), [Ws(s)];
    }
    case "NextIteration": {
      let s = S("tensor", e, t, n);
      return n.nextIteration(), [Ws(s)];
    }
    case "TensorArrayV3": {
      let s = S("size", e, t, n), r = S("dtype", e, t, n), a = S("elementShape", e, t, n), i = S("dynamicSize", e, t, n), o = S("clearAfterRead", e, t, n), u = S("identicalElementShapes", e, t, n), l = S("name", e, t, n), c = new g4(l, r, s, a, u, i, o);
      return n.addTensorArray(c), [c.idTensor, we(1)];
    }
    case "TensorArrayWriteV3": {
      let s = S("tensorArrayId", e, t, n), r = S("index", e, t, n), a = S("tensor", e, t, n), i = n.getTensorArray(s.id);
      return i.write(r, a), [i.idTensor];
    }
    case "TensorArrayReadV3": {
      let s = S("tensorArrayId", e, t, n), r = S("index", e, t, n);
      return [n.getTensorArray(s.id).read(r)];
    }
    case "TensorArrayGatherV3": {
      let s = S("tensorArrayId", e, t, n), r = S("indices", e, t, n), a = S("dtype", e, t, n);
      return [n.getTensorArray(s.id).gather(r, a)];
    }
    case "TensorArrayScatterV3": {
      let s = S("tensorArrayId", e, t, n), r = S("indices", e, t, n), a = S("tensor", e, t, n), i = n.getTensorArray(s.id);
      return i.scatter(r, a), [i.idTensor];
    }
    case "TensorArrayConcatV3": {
      let s = S("tensorArrayId", e, t, n), r = n.getTensorArray(s.id), a = S("dtype", e, t, n);
      return [r.concat(a)];
    }
    case "TensorArraySplitV3": {
      let s = S("tensorArrayId", e, t, n), r = S("tensor", e, t, n), a = S("lengths", e, t, n), i = n.getTensorArray(s.id);
      return i.split(a, r), [i.idTensor];
    }
    case "TensorArraySizeV3": {
      let s = S("tensorArrayId", e, t, n), r = n.getTensorArray(s.id);
      return [we(r.size(), "int32")];
    }
    case "TensorArrayCloseV3": {
      let s = S("tensorArrayId", e, t, n), r = n.getTensorArray(s.id);
      return r.clearAndClose(), [r.idTensor];
    }
    case "TensorListSetItem": {
      let s = S("tensorListId", e, t, n), r = S("index", e, t, n), a = S("tensor", e, t, n), i = n.getTensorList(s.id);
      return i.setItem(r, a), [i.idTensor];
    }
    case "TensorListGetItem": {
      let s = S("tensorListId", e, t, n), r = S("index", e, t, n), a = S("elementShape", e, t, n), i = S("elementDType", e, t, n);
      return [n.getTensorList(s.id).getItem(r, a, i)];
    }
    case "TensorListScatterV2":
    case "TensorListScatter": {
      let s = S("indices", e, t, n), r = S("tensor", e, t, n), a = S("elementShape", e, t, n), i = S("numElements", e, t, n), o = v4(r, s, a, i);
      return n.addTensorList(o), [o.idTensor];
    }
    case "TensorListReserve":
    case "EmptyTensorList": {
      let s = S("elementShape", e, t, n), r = S("elementDType", e, t, n), a;
      e.op === "TensorListReserve" ? a = "numElements" : a = "maxNumElements";
      let i = S(a, e, t, n), o = y4(s, r, i);
      return n.addTensorList(o), [o.idTensor];
    }
    case "TensorListGather": {
      let s = S("tensorListId", e, t, n), r = S("indices", e, t, n), a = S("elementShape", e, t, n), i = S("elementDType", e, t, n);
      return [n.getTensorList(s.id).gather(r, i, a)];
    }
    case "TensorListStack": {
      let s = S("tensorListId", e, t, n), r = S("elementShape", e, t, n), a = S("elementDType", e, t, n), i = S("numElements", e, t, n);
      return [n.getTensorList(s.id).stack(r, a, i)];
    }
    case "TensorListFromTensor": {
      let s = S("tensor", e, t, n), r = S("elementShape", e, t, n), a = S("elementDType", e, t, n), i = b4(s, r, a);
      return n.addTensorList(i), [i.idTensor];
    }
    case "TensorListConcat":
    case "TensorListConcatV2": {
      let s = S("tensorListId", e, t, n), r = n.getTensorList(s.id), a = S("dtype", e, t, n), i = S("elementShape", e, t, n);
      return [r.concat(a, i)];
    }
    case "TensorListPushBack": {
      let s = S("tensorListId", e, t, n), r = S("tensor", e, t, n), a = n.getTensorList(s.id);
      return a.pushBack(r), [a.idTensor];
    }
    case "TensorListPopBack": {
      let s = S("tensorListId", e, t, n), r = S("elementShape", e, t, n), a = S("elementDType", e, t, n);
      return [n.getTensorList(s.id).popBack(r, a)];
    }
    case "TensorListSplit": {
      let s = S("tensor", e, t, n), r = S("elementShape", e, t, n), a = S("lengths", e, t, n), i = x4(s, a, r);
      return n.addTensorList(i), [i.idTensor];
    }
    case "TensorListLength": {
      let s = S("tensorListId", e, t, n), r = n.getTensorList(s.id);
      return [we(r.size(), "int32")];
    }
    case "TensorListResize": {
      let s = S("tensorListId", e, t, n), r = S("size", e, t, n), i = n.getTensorList(s.id).resize(r);
      return n.addTensorList(i), [i.idTensor];
    }
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
function tw(e, t, n) {
  let [s, r] = S("fusedOps", e, t, n), a = s === "biasadd", i = !a, o = r === "prelu", u = s === "fusedbatchnorm", l = S("numArgs", e, t, n);
  if (a) {
    if (o && l !== 2)
      throw new Error("FusedConv2d and DepthwiseConv2d with BiasAdd and Prelu must have two extra arguments: bias and alpha.");
    if (!o && a && l !== 1)
      throw new Error("FusedConv2d and DepthwiseConv2d with BiasAdd must have one extra argument: bias.");
  }
  if (u)
    throw new Error("FusedConv2d and DepthwiseConv2d with FusedBatchNorm is not supported");
  let c = S("strides", e, t, n), p = id(e, t, n), d = S("dataFormat", e, t, n).toUpperCase(), h = S("dilations", e, t, n), [f, m] = S("args", e, t, n);
  i && (m = f, f = void 0);
  let g = S("leakyreluAlpha", e, t, n);
  return { stride: c, pad: p, dataFormat: d, dilations: h, biasArg: f, preluArg: m, activationFunc: r, leakyreluAlpha: g };
}
var k4 = (e, t, n) => {
  switch (e.op) {
    case "Conv1D": {
      let s = S("stride", e, t, n), r = S("pad", e, t, n), a = S("dataFormat", e, t, n).toUpperCase(), i = S("dilation", e, t, n);
      return [uS(S("x", e, t, n), S("filter", e, t, n), s, r, a, i)];
    }
    case "Conv2D": {
      let s = S("strides", e, t, n), r = id(e, t, n), a = S("dataFormat", e, t, n).toUpperCase(), i = S("dilations", e, t, n);
      return [da(S("x", e, t, n), S("filter", e, t, n), [s[1], s[2]], r, a, [i[1], i[2]])];
    }
    case "_FusedConv2D": {
      let { stride: s, pad: r, dataFormat: a, dilations: i, biasArg: o, preluArg: u, activationFunc: l, leakyreluAlpha: c } = tw(e, t, n);
      return [fa.conv2d({ x: S("x", e, t, n), filter: S("filter", e, t, n), strides: [s[1], s[2]], pad: r, dataFormat: a, dilations: [i[1], i[2]], bias: o, activation: l, preluActivationWeights: u, leakyreluAlpha: c })];
    }
    case "FusedDepthwiseConv2dNative": {
      let { stride: s, pad: r, dataFormat: a, dilations: i, biasArg: o, preluArg: u, activationFunc: l, leakyreluAlpha: c } = tw(e, t, n);
      return [fa.depthwiseConv2d({ x: S("x", e, t, n), filter: S("filter", e, t, n), strides: [s[1], s[2]], pad: r, dataFormat: a, dilations: [i[1], i[2]], bias: o, activation: l, preluActivationWeights: u, leakyreluAlpha: c })];
    }
    case "Conv2DBackpropInput":
    case "Conv2dTranspose": {
      let s = S("outputShape", e, t, n), r = S("strides", e, t, n), a = id(e, t, n);
      return [lS(S("x", e, t, n), S("filter", e, t, n), s, [r[1], r[2]], a)];
    }
    case "DepthwiseConv2dNative":
    case "DepthwiseConv2d": {
      let s = S("strides", e, t, n), r = id(e, t, n), a = S("dilations", e, t, n), i = S("dataFormat", e, t, n).toUpperCase();
      return [xp(S("input", e, t, n), S("filter", e, t, n), [s[1], s[2]], r, i, [a[1], a[2]])];
    }
    case "Conv3D": {
      let s = S("strides", e, t, n), r = S("pad", e, t, n), a = S("dataFormat", e, t, n).toUpperCase(), i = S("dilations", e, t, n);
      return [cS(S("x", e, t, n), S("filter", e, t, n), [s[1], s[2], s[3]], r, a, [i[1], i[2], i[3]])];
    }
    case "AvgPool": {
      let s = S("strides", e, t, n), r = S("pad", e, t, n), a = S("kernelSize", e, t, n);
      return [Qg(S("x", e, t, n), [a[1], a[2]], [s[1], s[2]], r)];
    }
    case "MaxPool": {
      let s = S("strides", e, t, n), r = S("pad", e, t, n), a = S("kernelSize", e, t, n);
      return [ub(S("x", e, t, n), [a[1], a[2]], [s[1], s[2]], r)];
    }
    case "MaxPoolWithArgmax": {
      let s = S("strides", e, t, n), r = S("pad", e, t, n), a = S("kernelSize", e, t, n), i = S("includeBatchInIndex", e, t, n), { result: o, indexes: u } = AD(S("x", e, t, n), [a[1], a[2]], [s[1], s[2]], r, i);
      return [o, u];
    }
    case "AvgPool3D": {
      let s = S("strides", e, t, n), r = S("pad", e, t, n), a = S("kernelSize", e, t, n);
      return [iS(S("x", e, t, n), [a[1], a[2], a[3]], [s[1], s[2], s[3]], r)];
    }
    case "MaxPool3D": {
      let s = S("strides", e, t, n), r = S("pad", e, t, n), a = S("kernelSize", e, t, n);
      return [SS(S("x", e, t, n), [a[1], a[2], a[3]], [s[1], s[2], s[3]], r)];
    }
    case "Dilation2D": {
      let s = S("strides", e, t, n), r = S("pad", e, t, n), a = S("dilations", e, t, n), i = s[1], o = s[2], u = a[1], l = a[2];
      return [wR(S("x", e, t, n), S("filter", e, t, n), [i, o], r, [u, l], "NHWC")];
    }
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var S4 = (e, t, n) => {
  switch (e.op) {
    case "Fill": {
      let s = S("shape", e, t, n), r = S("dtype", e, t, n), a = S("value", e, t, n);
      return [Ll(s, a, r)];
    }
    case "LinSpace": {
      let s = S("start", e, t, n), r = S("stop", e, t, n), a = S("num", e, t, n);
      return [uD(s, r, a)];
    }
    case "Multinomial": {
      let s = S("logits", e, t, n), r = S("numSamples", e, t, n), a = S("seed", e, t, n);
      return [VD(s, r, a)];
    }
    case "OneHot": {
      let s = S("indices", e, t, n), r = S("depth", e, t, n), a = S("onValue", e, t, n), i = S("offValue", e, t, n);
      return [Sd(s, r, a, i)];
    }
    case "Ones":
      return [Mn(S("shape", e, t, n), S("dtype", e, t, n))];
    case "OnesLike":
      return [Zn(S("x", e, t, n))];
    case "RandomUniform":
      return [Vl(S("shape", e, t, n), S("minval", e, t, n), S("maxval", e, t, n), S("dtype", e, t, n))];
    case "Range": {
      let s = S("start", e, t, n), r = S("stop", e, t, n), a = S("step", e, t, n);
      return [el(s, r, a, S("dtype", e, t, n))];
    }
    case "TruncatedNormal": {
      let s = S("shape", e, t, n), r = S("mean", e, t, n), a = S("stdDev", e, t, n), i = S("seed", e, t, n);
      return [vb(s, r, a, S("dtype", e, t, n), i)];
    }
    case "Zeros":
      return [$t(S("shape", e, t, n), S("dtype", e, t, n))];
    case "ZerosLike":
      return [je(S("x", e, t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
function Qf(e, t, n) {
  let s = S("boxes", e, t, n), r = S("scores", e, t, n), a = S("maxOutputSize", e, t, n), i = S("iouThreshold", e, t, n), o = S("scoreThreshold", e, t, n), u = S("softNmsSigma", e, t, n);
  return { boxes: s, scores: r, maxOutputSize: a, iouThreshold: i, scoreThreshold: o, softNmsSigma: u };
}
var I4 = async (e, t, n) => {
  switch (e.op) {
    case "NonMaxSuppressionV5": {
      let { boxes: s, scores: r, maxOutputSize: a, iouThreshold: i, scoreThreshold: o, softNmsSigma: u } = Qf(e, t, n), l = await jn.nonMaxSuppressionWithScoreAsync(s, r, a, i, o, u);
      return [l.selectedIndices, l.selectedScores];
    }
    case "NonMaxSuppressionV4": {
      let { boxes: s, scores: r, maxOutputSize: a, iouThreshold: i, scoreThreshold: o } = Qf(e, t, n), u = S("padToMaxOutputSize", e, t, n), l = await jn.nonMaxSuppressionPaddedAsync(s, r, a, i, o, u);
      return [l.selectedIndices, l.validOutputs];
    }
    case "NonMaxSuppressionV3":
    case "NonMaxSuppressionV2": {
      let { boxes: s, scores: r, maxOutputSize: a, iouThreshold: i, scoreThreshold: o } = Qf(e, t, n);
      return [await jn.nonMaxSuppressionAsync(s, r, a, i, o)];
    }
    case "Where": {
      let s = le(S("condition", e, t, n), "bool"), r = [await OS(s)];
      return s.dispose(), r;
    }
    case "ListDiff":
      return C3(S("x", e, t, n), S("y", e, t, n));
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var C4 = (e, t, n) => {
  switch (e.op) {
    case "LowerBound": {
      let s = S("sortedSequence", e, t, n), r = S("values", e, t, n);
      return [ND(s, r)];
    }
    case "TopKV2": {
      let s = S("x", e, t, n), r = S("k", e, t, n), a = S("sorted", e, t, n), i = X3(s, r, a);
      return [i.values, i.indices];
    }
    case "UpperBound": {
      let s = S("sortedSequence", e, t, n), r = S("values", e, t, n);
      return [tF(s, r)];
    }
    case "Unique": {
      let s = S("x", e, t, n), r = xx(s);
      return [r.values, r.indices];
    }
    case "UniqueV2": {
      let s = S("x", e, t, n), r = S("axis", e, t, n), a = xx(s, r);
      return [a.values, a.indices];
    }
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var N4 = (e, t, n) => {
  switch (e.op) {
    case "Const":
      return t[e.name];
    case "PlaceholderWithDefault":
      let s = S("default", e, t, n);
      return [un(e.name, t, n) || s];
    case "Placeholder":
      return [un(e.name, t, n)];
    case "Identity":
    case "StopGradient":
    case "FakeQuantWithMinMaxVars": {
      let l = S("x", e, t, n);
      return [Ws(l)];
    }
    case "IdentityN":
      return S("x", e, t, n).map((l) => Ws(l));
    case "Snapshot":
      let r = S("x", e, t, n);
      return [Ws(r)];
    case "Shape":
      return [Zt(S("x", e, t, n).shape, "int32")];
    case "ShapeN":
      return S("x", e, t, n).map((l) => Zt(l.shape));
    case "Size":
      return [we(S("x", e, t, n).size, "int32")];
    case "Rank":
      return [we(S("x", e, t, n).rank, "int32")];
    case "NoOp":
      return [we(1)];
    case "Print":
      let a = S("x", e, t, n), i = S("data", e, t, n), o = S("message", e, t, n), u = S("summarize", e, t, n);
      console.warn("The graph has a tf.print() operation,usually used for debugging, which slows down performance."), console.log(o);
      for (let l = 0; l < i.length; l++)
        console.log(Array.prototype.slice.call(i[l].dataSync()).slice(0, u));
      return [a];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var T4 = class {
  constructor(e, t) {
    this.keyDType = e, this.valueDType = t, this.handle = we(0), this.tensorMap = /* @__PURE__ */ new Map(), qt(this.handle);
  }
  get id() {
    return this.handle.id;
  }
  clearAndClose() {
    this.tensorMap.forEach((e) => e.dispose()), this.tensorMap.clear(), this.handle.dispose();
  }
  size() {
    return this.tensorMap.size;
  }
  tensorSize() {
    return we(this.size(), "int32");
  }
  async import(e, t) {
    this.checkKeyAndValueTensor(e, t);
    let n = await e.data();
    return this.tensorMap.forEach((s) => s.dispose()), this.tensorMap.clear(), j(() => {
      let s = Fs(t), r = n.length, a = s.length;
      w.assert(r === a, () => `The number of elements doesn't match, keys has ${r} elements, the values has ${a} elements.`);
      for (let i = 0; i < r; i++) {
        let o = n[i], u = s[i];
        qt(u), this.tensorMap.set(o, u);
      }
      return this.handle;
    });
  }
  async find(e, t) {
    this.checkKeyAndValueTensor(e, t);
    let n = await e.data();
    return j(() => {
      let s = [];
      for (let r = 0; r < n.length; r++) {
        let a = n[r], i = this.findWithDefault(a, t);
        s.push(i);
      }
      return es(s);
    });
  }
  findWithDefault(e, t) {
    let n = this.tensorMap.get(e);
    return n != null ? n : t;
  }
  checkKeyAndValueTensor(e, t) {
    if (e.dtype !== this.keyDType)
      throw new Error(`Expect key dtype ${this.keyDType}, but got ${e.dtype}`);
    if (t.dtype !== this.valueDType)
      throw new Error(`Expect value dtype ${this.valueDType}, but got ${t.dtype}`);
  }
};
var $4 = async (e, t, n, s) => {
  switch (e.op) {
    case "HashTable":
    case "HashTableV2": {
      let r = S("keyDType", e, t, n), a = S("valueDType", e, t, n), i = new T4(r, a);
      return s.addHashTable(e.name, i), [i.handle];
    }
    case "LookupTableImport":
    case "LookupTableImportV2": {
      let r = S("tableHandle", e, t, n, s), a = S("keys", e, t, n), i = S("values", e, t, n);
      return [await s.getHashTableById(r.id).import(a, i)];
    }
    case "LookupTableFind":
    case "LookupTableFindV2": {
      let r = S("tableHandle", e, t, n, s), a = S("keys", e, t, n), i = S("defaultValue", e, t, n);
      return [await s.getHashTableById(r.id).find(a, i)];
    }
    case "LookupTableSize":
    case "LookupTableSizeV2": {
      let r = S("tableHandle", e, t, n, s);
      return [s.getHashTableById(r.id).tensorSize()];
    }
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var _4 = (e, t, n) => {
  switch (e.op) {
    case "ResizeBilinear": {
      let s = S("images", e, t, n), r = S("size", e, t, n), a = S("alignCorners", e, t, n), i = S("halfPixelCenters", e, t, n);
      return [jn.resizeBilinear(s, [r[0], r[1]], a, i)];
    }
    case "ResizeNearestNeighbor": {
      let s = S("images", e, t, n), r = S("size", e, t, n), a = S("alignCorners", e, t, n), i = S("halfPixelCenters", e, t, n);
      return [jn.resizeNearestNeighbor(s, [r[0], r[1]], a, i)];
    }
    case "CropAndResize": {
      let s = S("image", e, t, n), r = S("boxes", e, t, n), a = S("boxInd", e, t, n), i = S("cropSize", e, t, n), o = S("method", e, t, n), u = S("extrapolationValue", e, t, n);
      return [jn.cropAndResize(s, r, a, i, o, u)];
    }
    case "ImageProjectiveTransformV3": {
      let s = S("images", e, t, n), r = S("transforms", e, t, n), a = S("outputShape", e, t, n), i = S("fillValue", e, t, n), o = S("interpolation", e, t, n), u = S("fillMode", e, t, n);
      return [jn.transform(s, r, o.toLowerCase(), u.toLowerCase(), i, a)];
    }
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var A4 = (e, t, n) => {
  switch (e.op) {
    case "Equal":
      return [Xn(S("a", e, t, n), S("b", e, t, n))];
    case "NotEqual":
      return [Ju(S("a", e, t, n), S("b", e, t, n))];
    case "Greater":
      return [Un(S("a", e, t, n), S("b", e, t, n))];
    case "GreaterEqual":
      return [Zo(S("a", e, t, n), S("b", e, t, n))];
    case "Less":
      return [vS(S("a", e, t, n), S("b", e, t, n))];
    case "LessEqual":
      return [Jo(S("a", e, t, n), S("b", e, t, n))];
    case "LogicalAnd":
      return [Ds(S("a", e, t, n), S("b", e, t, n))];
    case "LogicalNot":
      return [ob(S("a", e, t, n))];
    case "LogicalOr":
      return [wS(S("a", e, t, n), S("b", e, t, n))];
    case "Select":
    case "SelectV2":
      return [vn(S("condition", e, t, n), S("a", e, t, n), S("b", e, t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var E4 = (e, t, n) => {
  switch (e.op) {
    case "BatchMatMul":
    case "BatchMatMulV2":
    case "MatMul":
      return [Ve(S("a", e, t, n), S("b", e, t, n), S("transposeA", e, t, n), S("transposeB", e, t, n))];
    case "Einsum":
      return [_R(S("equation", e, t, n), ...S("tensors", e, t, n))];
    case "Transpose":
      return [Ge(S("x", e, t, n), S("perm", e, t, n))];
    case "_FusedMatMul":
      let [s, r] = S("fusedOps", e, t, n), a = s === "biasadd", i = r === "prelu", o = S("numArgs", e, t, n), u = S("leakyreluAlpha", e, t, n);
      if (a) {
        if (i && o !== 2)
          throw new Error("Fused MatMul with BiasAdd and Prelu must have two extra arguments: bias and alpha.");
        if (!i && o !== 1)
          throw new Error("Fused MatMul with BiasAdd must have one extra argument: bias.");
      }
      let [l, c] = S("args", e, t, n);
      return [fa.matMul({ a: S("a", e, t, n), b: S("b", e, t, n), transposeA: S("transposeA", e, t, n), transposeB: S("transposeB", e, t, n), bias: l, activation: r, preluActivationWeights: c, leakyreluAlpha: u })];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var R4 = (e, t, n) => {
  switch (e.op) {
    case "EuclideanNorm":
      return [UR(S("x", e, t, n), S("axis", e, t, n), S("keepDims", e, t, n))];
    case "FusedBatchNorm":
    case "FusedBatchNormV2":
      return [Qu(S("x", e, t, n), S("mean", e, t, n), S("variance", e, t, n), S("offset", e, t, n), S("scale", e, t, n), S("epsilon", e, t, n))];
    case "FusedBatchNormV3":
      return [Qu(S("x", e, t, n), S("mean", e, t, n), S("variance", e, t, n), S("offset", e, t, n), S("scale", e, t, n), S("epsilon", e, t, n))];
    case "LRN":
      return [cD(S("x", e, t, n), S("radius", e, t, n), S("bias", e, t, n), S("alpha", e, t, n), S("beta", e, t, n))];
    case "Softmax":
      return [gb(S("x", e, t, n))];
    case "LogSoftmax":
      return [xS(S("x", e, t, n))];
    case "SparseToDense":
      return [PS(S("sparseIndices", e, t, n), S("outputShape", e, t, n), S("sparseValues", e, t, n), S("defaultValue", e, t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var D4 = (e, t, n) => {
  switch (e.op) {
    case "Max": {
      let i = S("axis", e, t, n), o = S("keepDims", e, t, n);
      return [As(S("x", e, t, n), i, o)];
    }
    case "Mean": {
      let i = S("axis", e, t, n), o = S("keepDims", e, t, n);
      return [It(S("x", e, t, n), i, o)];
    }
    case "Min": {
      let i = S("axis", e, t, n), o = S("keepDims", e, t, n);
      return [wm(S("x", e, t, n), i, o)];
    }
    case "Sum": {
      let i = S("axis", e, t, n), o = S("keepDims", e, t, n);
      return [ve(S("x", e, t, n), i, o)];
    }
    case "All": {
      let i = S("axis", e, t, n), o = S("keepDims", e, t, n);
      return [nS(S("x", e, t, n), i, o)];
    }
    case "Any": {
      let i = S("axis", e, t, n), o = S("keepDims", e, t, n);
      return [ym(S("x", e, t, n), i, o)];
    }
    case "ArgMax": {
      let i = S("axis", e, t, n);
      return [Xu(S("x", e, t, n), i)];
    }
    case "ArgMin": {
      let i = S("axis", e, t, n);
      return [oE(S("x", e, t, n), i)];
    }
    case "Prod": {
      let i = S("axis", e, t, n), o = S("keepDims", e, t, n);
      return [IS(S("x", e, t, n), i, o)];
    }
    case "Cumprod": {
      let i = S("axis", e, t, n), o = S("exclusive", e, t, n), u = S("reverse", e, t, n);
      return [xm(S("x", e, t, n), i, o, u)];
    }
    case "Cumsum": {
      let i = S("axis", e, t, n), o = S("exclusive", e, t, n), u = S("reverse", e, t, n);
      return [hS(S("x", e, t, n), i, o, u)];
    }
    case "Bincount":
      let s = S("x", e, t, n), r = S("weights", e, t, n), a = S("size", e, t, n);
      return [oS(s, r, a)];
    case "DenseBincount": {
      let i = S("x", e, t, n), o = S("weights", e, t, n), u = S("size", e, t, n), l = S("binaryOutput", e, t, n);
      return [mR(i, o, u, l)];
    }
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var F4 = (e, t, n) => {
  switch (e.op) {
    case "ConcatV2":
    case "Concat": {
      let s = S("n", e, t, n), r = S("axis", e, t, n), a = S("tensors", e, t, n);
      return a = a.slice(0, s), [Ot(a, r)];
    }
    case "Gather": {
      let s = S("x", e, t, n), r = S("indices", e, t, n);
      return [Zu(s, le(r, "int32"), 0)];
    }
    case "GatherV2": {
      let s = S("axis", e, t, n), r = S("batchDims", e, t, n), a = S("x", e, t, n), i = S("indices", e, t, n);
      return [Zu(a, le(i, "int32"), s, r)];
    }
    case "Reverse": {
      let s = S("dims", e, t, n), r = [];
      for (let i = 0; i < s.length; i++)
        s[i] && r.push(i);
      let a = S("x", e, t, n);
      return [Jn(a, r)];
    }
    case "ReverseV2": {
      let s = S("axis", e, t, n), r = S("x", e, t, n);
      return [Jn(r, s)];
    }
    case "Slice": {
      let s = S("begin", e, t, n), r = S("size", e, t, n);
      return [qe(S("x", e, t, n), s, r)];
    }
    case "StridedSlice": {
      let s = S("begin", e, t, n), r = S("end", e, t, n), a = S("strides", e, t, n), i = S("beginMask", e, t, n), o = S("endMask", e, t, n), u = S("ellipsisMask", e, t, n), l = S("newAxisMask", e, t, n), c = S("shrinkAxisMask", e, t, n), p = S("x", e, t, n);
      return [H3(p, s, r, a, i, o, u, l, c)];
    }
    case "Pack":
      return j(() => {
        let s = S("axis", e, t, n), r = S("tensors", e, t, n), a = r[0].shape, i = mr(r[0]).shape, o = r.map((u) => {
          let l = w.arraysEqual(u.shape, a);
          if (!l && !w.arraysEqual(mr(u).shape, i))
            throw new Error("the input tensors shape does not match");
          return l ? u : U(u, a);
        });
        return [es(o, s)];
      });
    case "Unpack": {
      let s = S("axis", e, t, n), r = S("tensor", e, t, n);
      return Fs(r, s);
    }
    case "Tile": {
      let s = S("reps", e, t, n);
      return [hs(S("x", e, t, n), s)];
    }
    case "Split":
    case "SplitV": {
      let s = S("axis", e, t, n), r = S("numOrSizeSplits", e, t, n), a = S("x", e, t, n);
      return Bn(a, r, s);
    }
    case "ScatterNd": {
      let s = S("indices", e, t, n), r = S("values", e, t, n), a = S("shape", e, t, n);
      return [oF(s, r, a)];
    }
    case "GatherNd": {
      let s = S("x", e, t, n), r = S("indices", e, t, n);
      return [dF(s, r)];
    }
    case "SparseToDense": {
      let s = S("sparseIndices", e, t, n), r = S("outputShape", e, t, n), a = S("sparseValues", e, t, n), i = S("defaultValue", e, t, n);
      return [PS(s, a, r, a.dtype === i.dtype ? i : le(i, a.dtype))];
    }
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var O4 = (e, t, n) => {
  switch (e.op) {
    case "SparseFillEmptyRows": {
      let { outputIndices: s, outputValues: r, emptyRowIndicator: a, reverseIndexMap: i } = Hc.sparseFillEmptyRows(S("indices", e, t, n), S("values", e, t, n), S("denseShape", e, t, n), S("defaultValue", e, t, n));
      return [s, r, a, i];
    }
    case "SparseReshape": {
      let { outputIndices: s, outputShape: r } = Hc.sparseReshape(S("inputIndices", e, t, n), S("inputShape", e, t, n), S("newShape", e, t, n));
      return [s, r];
    }
    case "SparseSegmentMean":
      return [Hc.sparseSegmentMean(S("data", e, t, n), S("indices", e, t, n), S("segmentIds", e, t, n))];
    case "SparseSegmentSum":
      return [Hc.sparseSegmentSum(S("data", e, t, n), S("indices", e, t, n), S("segmentIds", e, t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var P4 = (e, t, n) => {
  switch (e.op) {
    case "FFT":
      return [bb(S("x", e, t, n))];
    case "IFFT":
      return [Td(S("x", e, t, n))];
    case "RFFT":
      return [yb(S("x", e, t, n))];
    case "IRFFT":
      return [RS(S("x", e, t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var z4 = (e, t, n) => {
  switch (e.op) {
    case "StringNGrams": {
      let { nGrams: s, nGramsSplits: r } = Hf.stringNGrams(S("data", e, t, n), S("dataSplits", e, t, n), S("separator", e, t, n), S("nGramWidths", e, t, n), S("leftPad", e, t, n), S("rightPad", e, t, n), S("padWidth", e, t, n), S("preserveShortSequences", e, t, n));
      return [s, r];
    }
    case "StringSplit": {
      let { indices: s, values: r, shape: a } = Hf.stringSplit(S("input", e, t, n), S("delimiter", e, t, n), S("skipEmpty", e, t, n));
      return [s, r, a];
    }
    case "StringToHashBucketFast":
      return [Hf.stringToHashBucketFast(S("input", e, t, n), S("numBuckets", e, t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
var M4 = (e, t, n) => {
  switch (e.op) {
    case "Cast":
      return [le(S("x", e, t, n), S("dtype", e, t, n))];
    case "ExpandDims": {
      let s = S("axis", e, t, n);
      return [Pn(S("x", e, t, n), s)];
    }
    case "Squeeze": {
      let s = S("axis", e, t, n);
      return [mr(S("x", e, t, n), s)];
    }
    case "Reshape":
      return [U(S("x", e, t, n), S("shape", e, t, n))];
    case "MirrorPad":
      return [OD(S("x", e, t, n), S("padding", e, t, n), S("mode", e, t, n))];
    case "PadV2":
    case "Pad":
      return [bi(S("x", e, t, n), S("padding", e, t, n), S("constantValue", e, t, n))];
    case "SpaceToBatchND": {
      let s = S("blockShape", e, t, n), r = S("paddings", e, t, n);
      return [cb(S("x", e, t, n), s, r)];
    }
    case "BatchToSpaceND": {
      let s = S("blockShape", e, t, n), r = S("crops", e, t, n);
      return [Zg(S("x", e, t, n), s, r)];
    }
    case "DepthToSpace": {
      let s = S("blockSize", e, t, n), r = S("dataFormat", e, t, n).toUpperCase();
      return [bR(S("x", e, t, n), s, r)];
    }
    case "BroadcastTo":
      return [ad(S("x", e, t, n), S("shape", e, t, n))];
    case "BroadcastArgs":
      return [UE(S("s0", e, t, n), S("s1", e, t, n))];
    default:
      throw TypeError(`Node type ${e.op} is not implemented`);
  }
};
function nw(e, t, n, s) {
  let r = ((a, i, o) => {
    switch (a.category) {
      case "arithmetic":
        return j(() => f4(a, i, o));
      case "basic_math":
        return j(() => m4(a, i, o));
      case "control":
        return w4(a, i, o);
      case "convolution":
        return j(() => k4(a, i, o));
      case "creation":
        return j(() => S4(a, i, o));
      case "dynamic":
        return I4(a, i, o);
      case "evaluation":
        return j(() => C4(a, i, o));
      case "image":
        return j(() => _4(a, i, o));
      case "graph":
        return j(() => N4(a, i, o));
      case "logical":
        return j(() => A4(a, i, o));
      case "matrices":
        return j(() => E4(a, i, o));
      case "normalization":
        return j(() => R4(a, i, o));
      case "reduction":
        return j(() => D4(a, i, o));
      case "slice_join":
        return j(() => F4(a, i, o));
      case "sparse":
        return j(() => O4(a, i, o));
      case "spectral":
        return j(() => P4(a, i, o));
      case "string":
        return j(() => z4(a, i, o));
      case "transformation":
        return j(() => M4(a, i, o));
      case "hash_table":
        return $4(a, i, o, s);
      case "custom":
        let u = i0(a.op);
        if (u && u.customExecutor)
          return u.customExecutor(new h4(a, i, o));
        throw TypeError(`Custom op ${a.op} is not registered.`);
      default:
        throw TypeError(`Unknown op '${a.op}'. File an issue at https://github.com/tensorflow/tfjs/issues so we can add it, or register a custom execution with tf.registerOp()`);
    }
  })(e, t, n);
  return w.isPromise(r) ? r.then((a) => [].concat(a)) : [].concat(r);
}
var sw = class {
  constructor(e = {}, t = {}, n = {}, s = {}) {
    this.weightMap = e, this.tensorArrayMap = t, this.tensorListMap = n, this.functionMap = s, this.rootContext = { id: 0, frameName: "", iterationId: 0 }, this.contexts = [this.rootContext], this.lastId = 0, this.generateCurrentContextIds();
  }
  newFrame(e, t) {
    return { id: e, frameName: t, iterationId: 0 };
  }
  set currentContext(e) {
    this.contexts !== e && (this.contexts = e, this.generateCurrentContextIds());
  }
  get currentContext() {
    return this.contexts;
  }
  get currentContextId() {
    return this._currentContextIds[0];
  }
  get currentContextIds() {
    return this._currentContextIds;
  }
  generateCurrentContextIds() {
    let e = [];
    for (let t = 0; t < this.contexts.length - 1; t++) {
      let n = this.contexts.slice(0, this.contexts.length - t);
      e.push(this.contextIdforContexts(n));
    }
    e.push(""), this._currentContextIds = e;
  }
  contextIdforContexts(e) {
    return e ? e.map((t) => t.id === 0 && t.iterationId === 0 ? "" : `${t.frameName}-${t.iterationId}`).join("/") : "";
  }
  enterFrame(e) {
    this.contexts && (this.lastId++, this.contexts = this.contexts.slice(), this.contexts.push(this.newFrame(this.lastId, e)), this._currentContextIds.unshift(this.contextIdforContexts(this.contexts)));
  }
  exitFrame() {
    if (this.contexts && this.contexts.length > 1)
      this.contexts = this.contexts.slice(), this.contexts.splice(-1), this.currentContextIds.shift();
    else
      throw new Error("Cannot exit frame, the context is empty");
  }
  nextIteration() {
    if (this.contexts && this.contexts.length > 0) {
      this.contexts = this.contexts.slice(), this.lastId++;
      let e = Object.assign({}, this.contexts[this.contexts.length - 1]);
      e.iterationId += 1, e.id = this.lastId, this.contexts.splice(-1, 1, e), this._currentContextIds.splice(0, 1, this.contextIdforContexts(this.contexts));
    } else
      throw new Error("Cannot increase frame iteration, the context is empty");
  }
  getWeight(e) {
    return this.weightMap[e];
  }
  addTensorArray(e) {
    this.tensorArrayMap[e.id] = e;
  }
  getTensorArray(e) {
    return this.tensorArrayMap[e];
  }
  addTensorList(e) {
    this.tensorListMap[e.id] = e;
  }
  getTensorList(e) {
    return this.tensorListMap[e];
  }
  dispose(e) {
    for (let t in this.tensorArrayMap)
      this.tensorArrayMap[t].clearAndClose(e);
    for (let t in this.tensorListMap)
      this.tensorListMap[t].clearAndClose(e);
  }
};
function rw(e, t, n, s) {
  let r = /* @__PURE__ */ new Set(), a = [], i = null, o = null, u = /* @__PURE__ */ new Set(), l = Object.keys(e).map((d) => _n(d)[0]), c = [];
  s != null && (c = s.map((d) => _n(d.name)[0]));
  let p = [...t];
  for (; p.length > 0; ) {
    let d = p.pop();
    if (($0(d) || U4(d) || G4(d)) && i == null && (i = d, o = i.children.map((h) => h.name).filter((h) => r.has(h))), r.add(d.name), n[d.name] == null && l.indexOf(d.name) === -1 && c.indexOf(d.name) === -1) {
      if (d.inputs.length === 0) {
        a.push(d.name);
        continue;
      }
      d.inputs.forEach((h) => {
        u.has(h.name) || (u.add(h.name), p.push(h));
      });
    }
  }
  return { inputs: e, outputs: t, usedNodes: r, missingInputs: a, dynamicNode: i, syncInputs: o };
}
function L4(e, t, n) {
  let { usedNodes: s, inputs: r } = n, a = [], i = Object.keys(r).map((c) => _n(c)[0]).map((c) => e.nodes[c]), o = e.initNodes;
  i.forEach((c) => {
    s.has(c.name) && a.push(c);
  }), e.weights.forEach((c) => {
    s.has(c.name) && a.push(c);
  }), o != null && o.forEach((c) => {
    s.has(c.name) && a.push(c);
  });
  let u = /* @__PURE__ */ new Set(), l = [];
  for (; a.length > 0; ) {
    let c = a.pop();
    u.add(c.name), t[c.name] || l.push(c), c.children.forEach((p) => {
      !u.has(p.name) && s.has(p.name) && p.inputs.every((d) => u.has(d.name)) && a.push(p);
    });
  }
  return l;
}
var B4 = ["Switch", "Merge", "Enter", "Exit", "NextIteration", "StatelessIf", "StatelessWhile", "if", "While"];
var V4 = ["NonMaxSuppressionV2", "NonMaxSuppressionV3", "NonMaxSuppressionV5", "Where"];
var W4 = ["HashTable", "HashTableV2", "LookupTableImport", "LookupTableImportV2", "LookupTableFind", "LookupTableFindV2", "LookupTableSize", "LookupTableSizeV2"];
function $0(e) {
  return B4.indexOf(e.op) >= 0;
}
function U4(e) {
  return V4.indexOf(e.op) >= 0;
}
function G4(e) {
  return W4.indexOf(e.op) >= 0;
}
var Gm = class {
  constructor(e, t) {
    this.graph = e, this.parent = t, this.compiledMap = /* @__PURE__ */ new Map(), this._weightMap = {}, this.SEPERATOR = ",", this._functions = {}, this._functionExecutorMap = {}, this.intermediateTensors = {}, this.keepTensorForDebug = false, this._outputs = e.outputs, this._inputs = e.inputs, this._initNodes = e.initNodes, this._signature = e.signature, this._functions = e.functions, e.functions != null && Object.keys(e.functions).forEach((n) => {
      this._functionExecutorMap[n] = new Gm(e.functions[n], this);
    });
  }
  get weightIds() {
    return this.parent ? this.parent.weightIds : this._weightIds;
  }
  get functionExecutorMap() {
    return this.parent ? this.parent.functionExecutorMap : this._functionExecutorMap;
  }
  get weightMap() {
    return this.parent ? this.parent.weightMap : this._weightMap;
  }
  set weightMap(e) {
    let t = Object.keys(e).map((n) => e[n].map((s) => s.id));
    this._weightIds = [].concat(...t), this._weightMap = e;
  }
  set resourceManager(e) {
    this._resourceManager = e;
  }
  get inputs() {
    return this._inputs.map((e) => ({ name: e.name, shape: e.attrParams.shape ? e.attrParams.shape.value : void 0, dtype: e.attrParams.dtype ? e.attrParams.dtype.value : void 0 }));
  }
  get outputs() {
    return this._outputs.map((e) => ({ name: e.name, shape: e.attrParams.shape ? e.attrParams.shape.value : void 0, dtype: e.attrParams.dtype ? e.attrParams.dtype.value : void 0 }));
  }
  get inputNodes() {
    return this._inputs.map((e) => e.signatureKey || e.name);
  }
  get outputNodes() {
    return this._outputs.map((e) => {
      let t = e.signatureKey || e.name;
      return e.defaultOutput ? `${t}:${e.defaultOutput}` : t;
    });
  }
  get functions() {
    return Object.keys(this._functions).reduce((e, t) => (e[t] = this._functions[t].signature, e), {});
  }
  getCompilationKey(e, t) {
    let n = e.map((r) => r.name).sort(), s = t.map((r) => r.name).sort();
    return n.join(this.SEPERATOR) + "--" + s.join(this.SEPERATOR);
  }
  compile(e, t) {
    let n = rw(e, t, this.weightMap, this._initNodes), { missingInputs: s, dynamicNode: r, syncInputs: a } = n;
    if (r != null)
      throw new Error(`This execution contains the node '${r.name}', which has the dynamic op '${r.op}'. Please use model.executeAsync() instead. Alternatively, to avoid the dynamic ops, specify the inputs [${a}]`);
    if (s.length > 0) {
      let i = t.map((u) => u.name), o = Object.keys(e);
      throw new Error(`Cannot compute the outputs [${i}] from the provided inputs [${o}]. Missing the following inputs: [${s}]`);
    }
    return L4(this.graph, this.weightMap, n);
  }
  execute(e, t) {
    e = this.mapInputs(e);
    let n = Object.keys(e).sort();
    this.checkInputs(e), this.checkInputShapeAndType(e), t = this.mapOutputs(t), this.checkOutputs(t);
    let s = n.map((c) => this.graph.nodes[_n(c)[0]]), r = t.map((c) => _n(c)[0]), a = r.map((c) => this.graph.nodes[c]);
    this.resetIntermediateTensors(), a.length === 0 && (a = this._outputs);
    let i = this.getCompilationKey(s, a), o = this.compiledMap.get(i);
    o == null && (o = this.compile(e, a), this.compiledMap.set(i, o));
    let u = {}, l = {};
    return j(() => {
      let c = new sw(this.weightMap, u, l, this.functionExecutorMap), p = { ...this.weightMap };
      Object.keys(e).forEach((f) => {
        let [m, g] = _n(f), b = [];
        b[g] = e[f], p[m] = b;
      });
      let d = this.getFrozenTensorIds(p), h = {};
      for (let f = 0; f < o.length; f++) {
        let m = o[f];
        if (!p[m.name]) {
          let g = nw(m, p, c, this._resourceManager);
          if (w.isPromise(g))
            throw new Error(`The execution of the op '${m.op}' returned a promise. Please use model.executeAsync() instead.`);
          p[m.name] = g, this.checkTensorForDisposal(m.name, m, p, c, d, r, h);
        }
      }
      return this.parent == null && c.dispose(d), t.map((f) => un(f, p, c));
    });
  }
  getFrozenTensorIds(e) {
    let t = [].concat.apply([], Object.keys(e).map((n) => e[n]).map((n) => n.map((s) => s.id)));
    return new Set(t);
  }
  checkTensorForDisposal(e, t, n, s, r, a, i) {
    t.category === "control" || a.indexOf(e) !== -1 || (n[e].forEach((o) => {
      o != null && (i[o.id] = (i[o.id] || 0) + t.children.length);
    }), t.inputs.forEach((o) => {
      if (o.category !== "control") {
        let u = qW(o.name, n, s);
        u != null && u.forEach((l) => {
          if (l && !l.kept && !r.has(l.id)) {
            let c = i[l.id];
            if (c === 1) {
              if (!this.keepTensorForDebug)
                l.dispose();
              else {
                let [p, d] = Ts(t.name, s);
                this.intermediateTensors[p] ? this.intermediateTensors[p][d] = l : (this.intermediateTensors[p] = [], this.intermediateTensors[p][d] = l);
              }
              delete i[l.id];
            } else
              c != null && i[l.id]--;
          }
        });
      }
    }));
  }
  async executeAsync(e, t) {
    return this._executeAsync(e, t);
  }
  disposeIntermediateTensors() {
    !this.intermediateTensors || (Object.keys(this.intermediateTensors).forEach((e) => this.intermediateTensors[e].forEach((t) => t.dispose())), this.disposeTensorsMap());
  }
  disposeTensorsMap() {
    !this.tensorsMap || Object.keys(this.tensorsMap).forEach((e) => {
      this.tensorsMap[e].forEach((n) => {
        n && !n.kept && !n.isDisposed && !this.keepIds.has(n.id) && n.dispose();
      });
    });
  }
  getIntermediateTensors() {
    return this.tensorsMap;
  }
  resetIntermediateTensors() {
    for (let e in this.intermediateTensors)
      this.intermediateTensors[e].forEach((t) => t.dispose()), delete this.intermediateTensors[e];
  }
  async _executeAsync(e, t, n = false, s = {}, r = {}) {
    n || (e = this.mapInputs(e), this.checkInputs(e), this.checkInputShapeAndType(e), t = this.mapOutputs(t), this.checkOutputs(t));
    try {
      this.keepTensorForDebug = K().getBool("KEEP_INTERMEDIATE_TENSORS");
    } catch (l) {
      console.warn(l.message);
    }
    this.resetIntermediateTensors();
    let a = new sw(this.weightMap, s, r, this.functionExecutorMap);
    this.tensorsMap = await this.executeWithControlFlow(e, a, t, n);
    let i = t.map((l) => un(l, this.tensorsMap, a)), o = i.map((l) => l.id), u = Object.keys(e).map((l) => e[l].id);
    return this.keepIds = /* @__PURE__ */ new Set([...o, ...u, ...this.weightIds]), this.keepTensorForDebug || this.disposeTensorsMap(), this.parent == null && a.dispose(this.keepIds), i;
  }
  async executeFunctionAsync(e, t, n) {
    let s = e.reduce((r, a, i) => (r[this.inputs[i].name] = a, r), {});
    return this._executeAsync(s, this.outputNodes, true, t, n);
  }
  async executeWithControlFlow(e, t, n, s) {
    let r = Object.keys(e), a = r.map((y) => this.graph.nodes[_n(y)[0]]), i = n.map((y) => _n(y)[0]), o = i.map((y) => this.graph.nodes[y]);
    o.length === 0 && (o = this._outputs);
    let { usedNodes: u, missingInputs: l, dynamicNode: c, syncInputs: p } = rw(e, o, this.weightMap, this._initNodes), d = [...a, ...this.graph.weights, ...this._initNodes || []].map((y) => ({ node: y, contexts: t.currentContext })), h = { ...this.weightMap };
    Object.keys(e).forEach((y) => {
      let [v, x] = _n(y), k = [];
      k[x] = e[y], h[v] = k;
    });
    let f = {}, m = this.getFrozenTensorIds(h), g = {};
    for (; d.length > 0; ) {
      let y = this.processStack(a, d, t, h, g, m, i, f, u);
      await Promise.all(y);
    }
    c == null && !s && console.warn("This model execution did not contain any nodes with control flow or dynamic output shapes. You can use model.execute() instead.");
    let b = o.filter((y) => !$0(y) && !un(y.name, h, t)).map((y) => y.name);
    if (b.length > 0) {
      let y = "";
      throw c != null && (y = `Alternatively, to avoid the dynamic ops, use model.execute() and specify the inputs [${p}]`), new Error(`Cannot compute the outputs [${b}] from the provided inputs [${r}]. Consider providing the following inputs: [${l}]. ${y}`);
    }
    return h;
  }
  processStack(e, t, n, s, r, a, i, o, u) {
    let l = [];
    for (; t.length > 0; ) {
      let c = t.pop();
      n.currentContext = c.contexts;
      let p = "";
      if (c.node.op === "Enter" && S("isConstant", c.node, s, n) && ([p] = Ts(c.node.name, n)), s[c.node.name] == null) {
        let d = nw(c.node, s, n, this._resourceManager);
        p || ([p] = Ts(c.node.name, n));
        let h = n.currentContext;
        w.isPromise(d) ? l.push(d.then((f) => (s[p] = f, n.currentContext = h, this.checkTensorForDisposal(p, c.node, s, n, a, i, o), this.processChildNodes(c.node, t, n, s, r, u), f))) : (s[p] = d, this.checkTensorForDisposal(p, c.node, s, n, a, i, o), this.processChildNodes(c.node, t, n, s, r, u));
      } else
        this.processChildNodes(c.node, t, n, s, r, u);
    }
    return l;
  }
  processChildNodes(e, t, n, s, r, a) {
    e.children.forEach((i) => {
      let [o] = Ts(i.name, n);
      r[o] || !a.has(i.name) || (i.op === "Merge" ? i.inputNames.some((u) => !!un(u, s, n)) && (r[o] = true, t.push({ contexts: n.currentContext, node: i })) : i.inputNames.every((u) => !!un(u, s, n)) && (r[o] = true, t.push({ contexts: n.currentContext, node: i })));
    });
  }
  dispose() {
    Object.keys(this.weightMap).forEach((e) => this.weightMap[e].forEach((t) => t.dispose()));
  }
  checkInputShapeAndType(e) {
    Object.keys(e).forEach((t) => {
      let n = e[t], [s] = _n(t), r = this.graph.nodes[s];
      if (r.attrParams.shape && r.attrParams.shape.value) {
        let a = r.attrParams.shape.value, i = a.length === n.shape.length && n.shape.every((o, u) => a[u] === -1 || a[u] === o);
        w.assert(i, () => `The shape of dict['${r.name}'] provided in model.execute(dict) must be [${a}], but was [${n.shape}]`);
      }
      r.attrParams.dtype && r.attrParams.dtype.value && w.assert(n.dtype === r.attrParams.dtype.value, () => `The dtype of dict['${r.name}'] provided in model.execute(dict) must be ${r.attrParams.dtype.value}, but was ${n.dtype}`);
    });
  }
  mapInputs(e) {
    let t = {};
    for (let n in e)
      if (this._signature != null && this._signature.inputs != null && this._signature.inputs[n] != null) {
        let s = this._signature.inputs[n];
        t[s.name] = e[n];
      } else
        t[n] = e[n];
    return t;
  }
  checkInputs(e) {
    let t = Object.keys(e).filter((n) => {
      let [s] = _n(n);
      return this.graph.nodes[s] == null;
    });
    if (t.length > 0)
      throw new Error(`The dict provided in model.execute(dict) has keys: [${t}] that are not part of graph`);
  }
  mapOutputs(e) {
    return e.map((t) => this._signature != null && this._signature.outputs != null && this._signature.outputs[t] != null ? this._signature.outputs[t].name : t, {});
  }
  checkOutputs(e) {
    e.forEach((t) => {
      let [n] = _n(t);
      if (!this.graph.nodes[n])
        throw new Error(`The output '${t}' is not found in the graph`);
    });
  }
};
var H4 = class {
  constructor(e = {}, t = {}) {
    this.hashTableNameToHandle = e, this.hashTableMap = t;
  }
  addHashTable(e, t) {
    this.hashTableNameToHandle[e] = t.handle, this.hashTableMap[t.id] = t;
  }
  getHashTableHandleByName(e) {
    return this.hashTableNameToHandle[e];
  }
  getHashTableById(e) {
    return this.hashTableMap[e];
  }
  dispose() {
    for (let e in this.hashTableMap)
      this.hashTableMap[e].clearAndClose(), delete this.hashTableMap[e];
    for (let e in this.hashTableNameToHandle)
      this.hashTableNameToHandle[e].dispose(), delete this.hashTableNameToHandle[e];
  }
};
var q4 = "?tfjs-format=file";
var j4 = "model.json";
var K4 = class {
  constructor(e, t = {}) {
    this.modelUrl = e, this.loadOptions = t, this.version = "n/a", t == null && (this.loadOptions = {}), this.resourceManager = new H4();
  }
  get modelVersion() {
    return this.version;
  }
  get inputNodes() {
    return this.executor.inputNodes;
  }
  get outputNodes() {
    return this.executor.outputNodes;
  }
  get inputs() {
    return this.executor.inputs;
  }
  get outputs() {
    return this.executor.outputs;
  }
  get weights() {
    return this.executor.weightMap;
  }
  get metadata() {
    return this.artifacts.userDefinedMetadata;
  }
  get modelSignature() {
    return this.signature;
  }
  findIOHandler() {
    let e = this.modelUrl;
    if (e.load != null)
      this.handler = e;
    else if (this.loadOptions.requestInit != null)
      this.handler = An.browserHTTPRequest(e, this.loadOptions);
    else {
      let t = An.getLoadHandlers(e, this.loadOptions);
      if (t.length === 0)
        t.push(An.browserHTTPRequest(e, this.loadOptions));
      else if (t.length > 1)
        throw new Error(`Found more than one (${t.length}) load handlers for URL '${[e]}'`);
      this.handler = t[0];
    }
  }
  async load() {
    if (this.findIOHandler(), this.handler.load == null)
      throw new Error("Cannot proceed with model loading because the IOHandler provided does not have the `load` method implemented.");
    let e = await this.handler.load();
    return this.loadSync(e);
  }
  loadSync(e) {
    this.artifacts = e;
    let t = this.artifacts.modelTopology, n;
    this.artifacts.userDefinedMetadata != null && this.artifacts.userDefinedMetadata.signature != null ? n = this.artifacts.userDefinedMetadata.signature : n = this.artifacts.signature, this.signature = n, this.version = `${t.versions.producer}.${t.versions.minConsumer}`;
    let s = An.decodeWeights(this.artifacts.weightData, this.artifacts.weightSpecs);
    if (this.executor = new Gm(Zx.Instance.transformGraph(t, this.signature)), this.executor.weightMap = this.convertTensorMapToTensorsMap(s), this.executor.resourceManager = this.resourceManager, e.modelInitializer != null && e.modelInitializer.node != null) {
      let r = Zx.Instance.transformGraph(e.modelInitializer);
      this.initializer = new Gm(r), this.initializer.weightMap = this.executor.weightMap, this.initializer.resourceManager = this.resourceManager, this.initializer.executeAsync({}, []);
    }
    return true;
  }
  async save(e, t) {
    if (typeof e == "string") {
      let n = An.getSaveHandlers(e);
      if (n.length === 0)
        throw new Error(`Cannot find any save handlers for URL '${e}'`);
      if (n.length > 1)
        throw new Error(`Found more than one (${n.length}) save handlers for URL '${e}'`);
      e = n[0];
    }
    if (e.save == null)
      throw new Error("GraphModel.save() cannot proceed because the IOHandler provided does not have the `save` attribute defined.");
    return e.save(this.artifacts);
  }
  predict(e, t) {
    return this.execute(e, this.outputNodes);
  }
  normalizeInputs(e) {
    if (!(e instanceof et) && !Array.isArray(e))
      return e;
    if (e = Array.isArray(e) ? e : [e], e.length !== this.inputNodes.length)
      throw new Error(`Input tensor count mismatch,the graph model has ${this.inputNodes.length} placeholders, while there are ${e.length} input tensors.`);
    return this.inputNodes.reduce((t, n, s) => (t[n] = e[s], t), {});
  }
  normalizeOutputs(e) {
    return e = e || this.outputNodes, Array.isArray(e) ? e : [e];
  }
  execute(e, t) {
    e = this.normalizeInputs(e), t = this.normalizeOutputs(t);
    let n = this.executor.execute(e, t);
    return n.length > 1 ? n : n[0];
  }
  async executeAsync(e, t) {
    e = this.normalizeInputs(e), t = this.normalizeOutputs(t);
    let n = await this.executor.executeAsync(e, t);
    return n.length > 1 ? n : n[0];
  }
  getIntermediateTensors() {
    return this.executor.getIntermediateTensors();
  }
  disposeIntermediateTensors() {
    this.executor.disposeIntermediateTensors();
  }
  convertTensorMapToTensorsMap(e) {
    return Object.keys(e).reduce((t, n) => (t[n] = [e[n]], t), {});
  }
  dispose() {
    this.executor.dispose(), this.initializer && this.initializer.dispose(), this.resourceManager.dispose();
  }
};
async function dhe(e, t = {}) {
  if (e == null)
    throw new Error("modelUrl in loadGraphModel() cannot be null. Please provide a url or an IOHandler that loads the model");
  t == null && (t = {}), t.fromTFHub && e.load == null && (e.endsWith("/") || (e = e + "/"), e = `${e}${j4}${q4}`);
  let n = new K4(e, t);
  return await n.load(), n;
}
var phe = "0.0.0";
var X4 = {};
Ee(X4, { CSVDataset: () => B0, Dataset: () => su, FileDataSource: () => j0, TextLineDataset: () => L0, URLDataSource: () => K0, array: () => yU, csv: () => _U, func: () => AU, generator: () => EU, microphone: () => DU, version_data: () => FU, webcam: () => RU, zip: () => vU });
var Y4 = wa(Xd());
var Q4 = wa(Xd());
function Z4(e, t) {
  return Md(e, t);
}
function Md(e, t, n = /* @__PURE__ */ new Map(), s = /* @__PURE__ */ new Set()) {
  if (e == null)
    return null;
  if (typeof Blob == "function" && e instanceof Blob)
    return e.slice();
  if (s.has(e))
    throw new Error("Circular references are not supported.");
  if (n.has(e))
    return n.get(e);
  let r = t(e);
  if (r.recurse && r.value !== null)
    throw new Error("A deep map function may not return both a value and recurse=true.");
  if (r.recurse)
    if (ao(e)) {
      let a = Array.isArray(e) ? [] : {};
      s.add(e);
      for (let i in e) {
        let o = e[i], u = Md(o, t, n, s);
        a[i] = u;
      }
      return s.delete(e), e.__proto__ && (a.__proto__ = e.__proto__), a;
    } else
      throw new Error(`Can't recurse into non-iterable type: ${e}`);
  else
    return n.set(e, r.value), r.value;
}
function J4(e, t = A0) {
  return _0(e, t);
}
function _0(e, t, n = /* @__PURE__ */ new Set()) {
  let s = e[0];
  if (n.has(s))
    throw new Error("Circular references are not supported.");
  let r = t(e);
  if (r.recurse && r.value !== null)
    throw new Error("A deep zip function may not return both a value and recurse=true.");
  if (r.recurse)
    if (ao(s)) {
      let a = Array.isArray(s) ? [] : {};
      n.add(s);
      for (let i in s) {
        let o = e.map((l) => l[i]), u = _0(o, t, n);
        a[i] = u;
      }
      return n.delete(s), a;
    } else
      throw new Error(`Can't recurse into non-iterable type: ${s}`);
  else
    return r.value;
}
function A0(e) {
  return e === null ? null : ao(e[0]) ? { value: null, recurse: true } : { value: e, recurse: false };
}
async function E0(e, t) {
  let n = /* @__PURE__ */ new Map();
  Md(e, t, n);
  for (let r of Array.from(n.keys())) {
    let a = n.get(r);
    if (w.isPromise(a)) {
      let i = await a;
      n.set(r, i);
    }
  }
  return Md(e, t, n);
}
function ao(e) {
  let t = false;
  if (K().get("IS_BROWSER"))
    t = e instanceof TextDecoder;
  else {
    let { StringDecoder: n } = Zw();
    t = e instanceof n;
  }
  return e != null && !ArrayBuffer.isView(e) && (Array.isArray(e) || typeof e == "object" && !(e instanceof et) && !(e instanceof Promise) && !t);
}
function eU(e) {
  return e == null || tU(e) || Array.isArray(e) || typeof e == "object" && e instanceof et || w.isTypedArray(e);
}
function tU(e) {
  return e === null || typeof e != "object" && typeof e != "function";
}
function nU(e) {
  return Z4(e, sU);
}
function sU(e) {
  return e instanceof et ? { value: e.clone(), recurse: false } : ao(e) ? { value: null, recurse: true } : { value: e, recurse: false };
}
var R0 = class {
  constructor(e) {
    if (this.capacity = e, this.begin = 0, this.end = 0, e == null)
      throw new RangeError("Can't create a ring buffer of unknown capacity.");
    if (e < 1)
      throw new RangeError("Can't create ring buffer of capacity < 1.");
    this.data = new Array(e), this.doubledCapacity = 2 * e;
  }
  wrap(e) {
    for (; e < 0; )
      e += this.doubledCapacity;
    return e % this.doubledCapacity;
  }
  get(e) {
    if (e < 0)
      throw new RangeError("Can't get item at a negative index.");
    return this.data[e % this.capacity];
  }
  set(e, t) {
    if (e < 0)
      throw new RangeError("Can't set item at a negative index.");
    this.data[e % this.capacity] = t;
  }
  length() {
    let e = this.end - this.begin;
    return e < 0 && (e = this.doubledCapacity + e), e;
  }
  isFull() {
    return this.length() === this.capacity;
  }
  isEmpty() {
    return this.length() === 0;
  }
  push(e) {
    if (this.isFull())
      throw new RangeError("Ring buffer is full.");
    this.set(this.end, e), this.end = this.wrap(this.end + 1);
  }
  pushAll(e) {
    for (let t of e)
      this.push(t);
  }
  pop() {
    if (this.isEmpty())
      throw new RangeError("Ring buffer is empty.");
    this.end = this.wrap(this.end - 1);
    let e = this.get(this.end);
    return this.set(this.end, void 0), e;
  }
  unshift(e) {
    if (this.isFull())
      throw new RangeError("Ring buffer is full.");
    this.begin = this.wrap(this.begin - 1), this.set(this.begin, e);
  }
  shift() {
    if (this.isEmpty())
      throw new RangeError("Ring buffer is empty.");
    let e = this.get(this.begin);
    return this.set(this.begin, void 0), this.begin = this.wrap(this.begin + 1), e;
  }
  shuffleExcise(e) {
    if (this.isEmpty())
      throw new RangeError("Ring buffer is empty.");
    let t = this.wrap(this.begin + e), n = this.get(t);
    return this.set(t, this.pop()), n;
  }
};
var D0 = class extends R0 {
  constructor() {
    super(D0.INITIAL_CAPACITY);
  }
  isFull() {
    return false;
  }
  push(e) {
    super.isFull() && this.expand(), super.push(e);
  }
  unshift(e) {
    super.isFull() && this.expand(), super.unshift(e);
  }
  expand() {
    let e = this.capacity * 2, t = new Array(e), n = this.length();
    for (let s = 0; s < n; s++)
      t[s] = this.get(this.wrap(this.begin + s));
    this.data = t, this.capacity = e, this.doubledCapacity = 2 * this.capacity, this.begin = 0, this.end = n;
  }
};
var F0 = D0;
F0.INITIAL_CAPACITY = 32;
function O0(e) {
  return new iU(e);
}
function rv(e) {
  return new oU(e);
}
function rU(e, t) {
  return new P0(e, t);
}
function aU(e, t = z0.FAIL) {
  return new gU(e, t);
}
var Gt = class {
  async toArray() {
    let e = [], t = await this.next();
    for (; !t.done; )
      e.push(t.value), t = await this.next();
    return e;
  }
  async toArrayForTest() {
    let e = this.prefetch(100), t = [], n = await e.next();
    for (; !n.done; )
      t.push(n.value), n = await e.next();
    return t;
  }
  async resolveFully() {
    let e = await this.next();
    for (; !e.done; )
      e = await this.next();
  }
  async resolveWhile(e) {
    let t = await this.next(), n = e(t.value);
    for (; !t.done && n; )
      t = await this.next(), n = e(t.value);
  }
  handleErrors(e) {
    return new fU(this, e);
  }
  filter(e) {
    return new pU(this, e);
  }
  map(e) {
    return new hU(this, e);
  }
  mapAsync(e) {
    return new aw(this, e);
  }
  serialMapAsync(e) {
    return new aw(this, e).serial();
  }
  flatmap(e) {
    return new mU(this, e);
  }
  async forEachAsync(e) {
    return this.map(e).resolveFully();
  }
  async serialForEach(e) {
    return this.serialMapAsync(e).resolveWhile((t) => t === true);
  }
  rowMajorBatch(e, t = true) {
    return new dU(this, e, t);
  }
  columnMajorBatch(e, t = true, n = A0) {
    return this.rowMajorBatch(e, t).map((r) => J4(r, n));
  }
  concatenate(e, t) {
    return new P0(O0([this, e]), t);
  }
  take(e) {
    return e < 0 || e == null ? this : new cU(this, e);
  }
  skip(e) {
    return e < 0 || e == null ? this : new lU(this, e);
  }
  prefetch(e) {
    return new M0(this, e);
  }
  shuffle(e, t) {
    return new bU(this, e, t);
  }
  serial() {
    return new uU(this);
  }
};
var iU = class extends Gt {
  constructor(e) {
    super(), this.items = e, this.trav = 0;
  }
  summary() {
    return `Array of ${this.items.length} items`;
  }
  async next() {
    if (this.trav >= this.items.length)
      return { value: null, done: true };
    let e = this.items[this.trav];
    return this.trav++, { value: nU(e), done: false };
  }
};
var oU = class extends Gt {
  constructor(e) {
    super(), this.nextFn = e;
  }
  summary() {
    return "Function call";
  }
  async next() {
    try {
      return this.nextFn();
    } catch (e) {
      throw e.message = `Error thrown while iterating through a dataset: ${e.message}`, e;
    }
  }
};
var uU = class extends Gt {
  constructor(e) {
    super(), this.upstream = e, this.lastRead = Promise.resolve({ value: null, done: false });
  }
  summary() {
    return `${this.upstream.summary()} -> Serial`;
  }
  async next() {
    return this.lastRead = this.lastRead.then(() => this.serialNext()), this.lastRead;
  }
  async serialNext() {
    return this.upstream.next();
  }
};
var lU = class extends Gt {
  constructor(e, t) {
    super(), this.upstream = e, this.maxCount = t, this.count = 0, this.lastRead = Promise.resolve({ value: null, done: false });
  }
  summary() {
    return `${this.upstream.summary()} -> Skip`;
  }
  async next() {
    return this.lastRead = this.lastRead.then(() => this.serialNext()), this.lastRead;
  }
  async serialNext() {
    for (; this.count++ < this.maxCount; ) {
      let e = await this.upstream.next();
      if (e.done)
        return e;
      De(e.value);
    }
    return this.upstream.next();
  }
};
var cU = class extends Gt {
  constructor(e, t) {
    super(), this.upstream = e, this.maxCount = t, this.count = 0;
  }
  summary() {
    return `${this.upstream.summary()} -> Take`;
  }
  async next() {
    return this.count++ >= this.maxCount ? { value: null, done: true } : this.upstream.next();
  }
};
var dU = class extends Gt {
  constructor(e, t, n = true) {
    super(), this.upstream = e, this.batchSize = t, this.enableSmallLastBatch = n, this.lastRead = Promise.resolve({ value: null, done: false });
  }
  summary() {
    return `${this.upstream.summary()} -> RowMajorBatch`;
  }
  async next() {
    return this.lastRead = this.lastRead.then(() => this.serialNext()), this.lastRead;
  }
  async serialNext() {
    let e = [];
    for (; e.length < this.batchSize; ) {
      let t = await this.upstream.next();
      if (t.done)
        return this.enableSmallLastBatch && e.length > 0 ? { value: e, done: false } : { value: null, done: true };
      e.push(t.value);
    }
    return { value: e, done: false };
  }
};
var pU = class extends Gt {
  constructor(e, t) {
    super(), this.upstream = e, this.predicate = t, this.lastRead = Promise.resolve({ value: null, done: false });
  }
  summary() {
    return `${this.upstream.summary()} -> Filter`;
  }
  async next() {
    return this.lastRead = this.lastRead.then(() => this.serialNext()), this.lastRead;
  }
  async serialNext() {
    for (; ; ) {
      let e = await this.upstream.next();
      if (e.done || this.predicate(e.value))
        return e;
      De(e.value);
    }
  }
};
var hU = class extends Gt {
  constructor(e, t) {
    super(), this.upstream = e, this.transform = t;
  }
  summary() {
    return `${this.upstream.summary()} -> Map`;
  }
  async next() {
    let e = await this.upstream.next();
    if (e.done)
      return { value: null, done: true };
    let t = _s.getTensorsInContainer(e.value), n = this.transform(e.value), s = _s.getTensorsInContainer(n);
    for (let r of t)
      _s.isTensorInList(r, s) || r.dispose();
    return { value: n, done: false };
  }
};
var fU = class extends Gt {
  constructor(e, t) {
    super(), this.upstream = e, this.handler = t, this.count = 0, this.lastRead = Promise.resolve({ value: null, done: false });
  }
  summary() {
    return `${this.upstream.summary()} -> handleErrors`;
  }
  async next() {
    return this.lastRead = this.lastRead.then(() => this.serialNext()), this.lastRead;
  }
  async serialNext() {
    for (; ; )
      try {
        return await this.upstream.next();
      } catch (e) {
        if (!this.handler(e))
          return { value: null, done: true };
      }
  }
};
var aw = class extends Gt {
  constructor(e, t) {
    super(), this.upstream = e, this.transform = t;
  }
  summary() {
    return `${this.upstream.summary()} -> AsyncMap`;
  }
  async next() {
    let e = await this.upstream.next();
    if (e.done)
      return { value: null, done: true };
    let t = _s.getTensorsInContainer(e.value), n = await this.transform(e.value), s = _s.getTensorsInContainer(n);
    for (let r of t)
      _s.isTensorInList(r, s) || r.dispose();
    return { value: n, done: false };
  }
};
var av = class extends Gt {
  constructor() {
    super(), this.outputQueue = new F0(), this.lastRead = Promise.resolve({ value: null, done: false });
  }
  async next() {
    return this.lastRead = this.lastRead.then(() => this.serialNext()), this.lastRead;
  }
  async serialNext() {
    for (; this.outputQueue.length() === 0; )
      if (!await this.pump())
        return { value: null, done: true };
    return { value: this.outputQueue.shift(), done: false };
  }
};
var mU = class extends av {
  constructor(e, t) {
    super(), this.upstream = e, this.transform = t;
  }
  summary() {
    return `${this.upstream.summary()} -> Flatmap`;
  }
  async pump() {
    let e = await this.upstream.next();
    if (e.done)
      return false;
    let t = _s.getTensorsInContainer(e.value), n = this.transform(e.value), s = _s.getTensorsInContainer(n);
    this.outputQueue.pushAll(n);
    for (let r of t)
      _s.isTensorInList(r, s) || r.dispose();
    return true;
  }
};
var P0 = class extends Gt {
  constructor(e, t) {
    super(), this.baseErrorHandler = t, this.lastRead = null, this.iterator = null, this.moreIterators = e;
  }
  summary() {
    return "TODO: fill in upstream of chained summaries -> Chained";
  }
  async next() {
    return this.lastRead = this.readFromChain(this.lastRead), this.lastRead;
  }
  async readFromChain(e) {
    if (await e, this.iterator == null) {
      let n = await this.moreIterators.next();
      if (n.done)
        return { value: null, done: true };
      this.iterator = n.value, this.baseErrorHandler != null && (this.iterator = this.iterator.handleErrors(this.baseErrorHandler));
    }
    let t = await this.iterator.next();
    return t.done ? (this.iterator = null, this.readFromChain(e)) : t;
  }
};
var z0 = ((e) => (e[e.FAIL = 0] = "FAIL", e[e.SHORTEST = 1] = "SHORTEST", e[e.LONGEST = 2] = "LONGEST", e))(z0 || {});
var gU = class extends Gt {
  constructor(e, t = 0) {
    super(), this.iterators = e, this.mismatchMode = t, this.count = 0, this.currentPromise = null;
  }
  summary() {
    return "{TODO: fill in upstream of zip summaries} -> Zip";
  }
  async nextState(e) {
    await e;
    let t = 0, n = 0;
    function s(a) {
      return a instanceof Gt ? { value: a.next().then((o) => (t++, o.done && n++, o.value)), recurse: false } : { value: null, recurse: true };
    }
    let r = await E0(this.iterators, s);
    if (t === n)
      return { value: null, done: true };
    if (n > 0)
      switch (this.mismatchMode) {
        case 0:
          throw new Error(`Zipped streams should have the same length. Mismatched at element ${this.count}.`);
        case 1:
          return { value: null, done: true };
        case 2:
        default:
      }
    return this.count++, { value: r, done: false };
  }
  async next() {
    return this.currentPromise = this.nextState(this.currentPromise), this.currentPromise;
  }
};
var M0 = class extends Gt {
  constructor(e, t) {
    super(), this.upstream = e, this.bufferSize = t, this.buffer = new R0(t);
  }
  summary() {
    return `${this.upstream.summary()} -> Prefetch`;
  }
  refill() {
    for (; !this.buffer.isFull(); ) {
      let e = this.upstream.next();
      this.buffer.push(e);
    }
  }
  next() {
    return this.refill(), this.buffer.shift();
  }
};
var bU = class extends M0 {
  constructor(e, t, n) {
    super(e, t), this.upstream = e, this.windowSize = t, this.upstreamExhausted = false, this.random = Q4.alea(n || w.now().toString()), this.lastRead = Promise.resolve({ value: null, done: false });
  }
  async next() {
    return this.lastRead = this.lastRead.then(() => this.serialNext()), this.lastRead;
  }
  randomInt(e) {
    return Math.floor(this.random() * e);
  }
  chooseIndex() {
    return this.randomInt(this.buffer.length());
  }
  async serialNext() {
    for (this.upstreamExhausted || this.refill(); !this.buffer.isEmpty(); ) {
      let e = this.chooseIndex(), t = await this.buffer.shuffleExcise(e);
      if (t.done)
        this.upstreamExhausted = true;
      else
        return this.refill(), t;
    }
    return { value: null, done: true };
  }
};
var su = class {
  constructor() {
    this.size = null;
  }
  batch(e, t = true) {
    let n = this;
    w.assert(e > 0, () => `batchSize needs to be positive, but it is
      ${e}`);
    let s;
    return this.size === 1 / 0 || this.size == null ? s = this.size : t ? s = Math.ceil(this.size / e) : s = Math.floor(this.size / e), $n(async () => (await n.iterator()).columnMajorBatch(e, t, xU), s);
  }
  concatenate(e) {
    let t = this, n;
    return this.size === 1 / 0 || e.size === 1 / 0 ? n = 1 / 0 : this.size != null && e.size != null ? n = this.size + e.size : n = null, $n(async () => (await t.iterator()).concatenate(await e.iterator()), n);
  }
  filter(e) {
    let t = this, n;
    return this.size === 1 / 0 ? n = 1 / 0 : n = null, $n(async () => (await t.iterator()).filter((s) => j(() => e(s))), n);
  }
  async forEachAsync(e) {
    return (await this.iterator()).forEachAsync(e);
  }
  map(e) {
    let t = this;
    return $n(async () => (await t.iterator()).map((n) => j(() => e(n))), this.size);
  }
  mapAsync(e) {
    let t = this;
    return $n(async () => (await t.iterator()).mapAsync(e), this.size);
  }
  prefetch(e) {
    if (e == null)
      throw new RangeError("`Dataset.prefetch()` requires bufferSize to be specified.");
    let t = this;
    return $n(async () => (await t.iterator()).prefetch(e), this.size);
  }
  repeat(e) {
    let t = this, n;
    return this.size != null && e > 0 ? n = this.size * e : e === 0 ? n = 0 : this.size != null && (e === void 0 || e < 0) ? n = 1 / 0 : n = null, $n(async () => {
      let s = rv(async () => ({ value: await t.iterator(), done: false }));
      return rU(s.take(e));
    }, n);
  }
  skip(e) {
    let t = this, n;
    return this.size != null && e >= 0 && this.size >= e ? n = this.size - e : this.size != null && (this.size < e || e === void 0 || e < 0) ? n = 0 : n = null, $n(async () => (await t.iterator()).skip(e), n);
  }
  shuffle(e, t, n = true) {
    if (e == null || e < 0)
      throw this.size == null ? new RangeError("`Dataset.shuffle()` requires bufferSize to be specified.") : new RangeError(`\`Dataset.shuffle()\` requires bufferSize to be specified.  If your data fits in main memory (for regular JS objects), and/or GPU memory (for \`tf.Tensor\`s), consider setting bufferSize to the dataset size (${this.size} elements)`);
    let s = this, r = Y4.alea(t || w.now().toString());
    return $n(async () => {
      let a = r.int32();
      return n && (a += r.int32()), (await s.iterator()).shuffle(e, a.toString());
    }, this.size);
  }
  take(e) {
    let t = this, n;
    return this.size != null && this.size > e ? n = e : this.size != null && this.size <= e ? n = this.size : n = null, $n(async () => (await t.iterator()).take(e), n);
  }
  async toArray() {
    if (this.size === 1 / 0)
      throw new Error("Can not convert infinite data stream to array.");
    return (await this.iterator()).toArray();
  }
  async toArrayForTest() {
    if (this.size === 1 / 0)
      throw new Error("Can not convert infinite data stream to array.");
    return (await this.iterator()).toArrayForTest();
  }
};
su.MAX_BUFFER_SIZE = 1e4;
function $n(e, t = null) {
  return new class extends su {
    constructor() {
      super(...arguments), this.size = t;
    }
    async iterator() {
      return e();
    }
  }();
}
function yU(e) {
  return $n(async () => O0(e), e.length);
}
function vU(e) {
  if (!ao(e))
    throw new Error("The argument to zip() must be an object or array.");
  let t;
  if (Array.isArray(e))
    for (let n = 0; n < e.length; n++)
      t = t == null ? e[n].size : Math.min(t, e[n].size);
  else if (e instanceof Object)
    for (let n in e)
      t = t == null ? e[n].size : Math.min(t, e[n].size);
  return $n(async () => {
    let n = await E0(e, (s) => {
      if (s instanceof su)
        return { value: s.iterator(), recurse: false };
      if (ao(s))
        return { value: null, recurse: true };
      throw new Error("Leaves of the structure passed to zip() must be Datasets, not primitives.");
    });
    return aU(n, 1);
  }, t);
}
function xU(e) {
  if (e === null)
    return null;
  let t = e[0];
  return eU(t) ? { value: wU(e), recurse: false } : { value: null, recurse: true };
}
function wU(e) {
  if (e.length === 0)
    throw new Error("Can't make a batch of zero elements.");
  return e[0] instanceof et ? es(e) : ms(e);
}
var L0 = class extends su {
  constructor(e) {
    super(), this.input = e;
  }
  async iterator() {
    return (await this.input.iterator()).decodeUTF8().split(`
`).map((s) => (s.endsWith("\r") && (s = s.slice(0, -1)), s));
  }
};
var Qc = '"';
var Eu = Symbol("out");
var iw = Symbol("field");
var Zc = Symbol("quote");
var Zf = Symbol("quoteafterquote");
var ow = Symbol("quoteinquote");
var B0 = class extends su {
  constructor(e, t) {
    super(), this.input = e, this.hasHeader = true, this.fullColumnNames = null, this.columnNamesValidated = false, this.columnConfigs = null, this.configuredColumnsOnly = false, this.delimiter = ",", this.delimWhitespace = false, this.base = new L0(e), t || (t = {}), this.hasHeader = t.hasHeader !== false, this.fullColumnNames = t.columnNames, this.columnConfigs = t.columnConfigs, this.configuredColumnsOnly = t.configuredColumnsOnly, t.delimWhitespace ? (w.assert(t.delimiter == null, () => "Delimiter should not be provided when delimWhitespace is true."), this.delimWhitespace = true, this.delimiter = " ") : this.delimiter = t.delimiter ? t.delimiter : ",";
  }
  async columnNames() {
    return this.columnNamesValidated || await this.setColumnNames(), this.configuredColumnsOnly ? Object.keys(this.columnConfigs) : this.fullColumnNames;
  }
  async setColumnNames() {
    let e = await this.maybeReadHeaderLine();
    if (!this.fullColumnNames && !e)
      throw new Error("Column names must be provided if there is no header line.");
    this.fullColumnNames && e && w.assert(e.length === this.fullColumnNames.length, () => "The length of provided columnNames (" + this.fullColumnNames.length.toString() + ") does not match the length of the header line read from file (" + e.length.toString() + ")."), this.fullColumnNames || (this.fullColumnNames = e);
    let t = this.fullColumnNames.reduce((s, r) => (s[r] = s[r] + 1 || 1, s), {}), n = Object.keys(t).filter((s) => t[s] > 1);
    if (w.assert(n.length === 0, () => "Duplicate column names found: " + n.toString()), this.columnConfigs) {
      for (let s of Object.keys(this.columnConfigs))
        if (this.fullColumnNames.indexOf(s) === -1)
          throw new Error('The key "' + s + '" provided in columnConfigs does not match any of the column names (' + this.fullColumnNames.toString() + ").");
    }
    this.columnNamesValidated = true;
  }
  async maybeReadHeaderLine() {
    if (this.hasHeader) {
      let t = await (await this.base.iterator()).next();
      if (t.done)
        throw new Error("No data was found for CSV parsing.");
      let n = t.value;
      return this.parseRow(n, false);
    } else
      return null;
  }
  async iterator() {
    this.columnNamesValidated || await this.setColumnNames();
    let e = await this.base.iterator();
    return this.hasHeader && (e = e.skip(1)), e.map((t) => this.makeDataElement(t));
  }
  makeDataElement(e) {
    let t = this.parseRow(e), n = {}, s = {};
    for (let r = 0; r < this.fullColumnNames.length; r++) {
      let a = this.fullColumnNames[r], i = this.columnConfigs ? this.columnConfigs[a] : null;
      if (!(this.configuredColumnsOnly && !i)) {
        let o = t[r], u = null;
        if (o === "")
          if (i && i.default !== void 0)
            u = i.default;
          else {
            if (i && (i.required || i.isLabel))
              throw new Error(`Required column ${a} is empty in this line: ${e}`);
            u = void 0;
          }
        else {
          let l = Number(o);
          if (isNaN(l))
            i && i.dtype === "bool" ? u = this.getBoolean(o) : u = o;
          else if (!i || !i.dtype)
            u = l;
          else
            switch (i.dtype) {
              case "float32":
                u = l;
                break;
              case "int32":
                u = Math.floor(l);
                break;
              case "bool":
                u = this.getBoolean(o);
                break;
              default:
                u = l;
            }
        }
        i && i.isLabel ? s[a] = u : n[a] = u;
      }
    }
    return Object.keys(s).length === 0 ? n : { xs: n, ys: s };
  }
  getBoolean(e) {
    return e === "1" || e.toLowerCase() === "true" ? 1 : 0;
  }
  parseRow(e, t = true) {
    let n = [], s = 0, r = e.length, a = Eu;
    for (let i = 0; i < r; i++)
      switch (a) {
        case Eu:
          switch (e.charAt(i)) {
            case Qc:
              s = i + 1, a = Zc;
              break;
            case this.delimiter:
              if (s = i + 1, this.delimiter === " " && this.delimWhitespace)
                break;
              n.push(""), a = Eu;
              break;
            default:
              a = iw, s = i;
              break;
          }
          break;
        case iw:
          switch (e.charAt(i)) {
            case this.delimiter:
              n.push(e.substring(s, i)), a = Eu, s = i + 1;
              break;
            default:
          }
          break;
        case Zc:
          switch (e.charAt(i)) {
            case Qc:
              a = Zf;
              break;
            default:
          }
          break;
        case Zf:
          switch (e.charAt(i)) {
            case this.delimiter:
              n.push(e.substring(s, i - 1)), a = Eu, s = i + 1;
              break;
            case Qc:
              a = Zc;
              break;
            default:
              a = ow;
              break;
          }
          break;
        case ow:
          switch (e.charAt(i)) {
            case Qc:
              a = Zc;
              break;
            default:
          }
          break;
        default:
      }
    if (a === Zf ? n.push(e.substring(s, r - 1)) : n.push(e.substring(s)), t && n.length !== this.fullColumnNames.length)
      throw new Error(`Invalid row in csv file. Should have ${this.fullColumnNames.length} elements in a row, but got ${n}`);
    return n;
  }
};
var V0 = class extends Gt {
  constructor(e) {
    super(), this.microphoneConfig = e, this.isClosed = false, this.fftSize = e.fftSize || 1024;
    let t = Math.log2(this.fftSize);
    if (this.fftSize < 0 || t < 4 || t > 14 || !Number.isInteger(t))
      throw new Error(`Invalid fftSize: it must be a power of 2 between 2 to 4 and 2 to 14, but got ${this.fftSize}`);
    if (this.numFrames = e.numFramesPerSpectrogram || 43, this.sampleRateHz = e.sampleRateHz, this.columnTruncateLength = e.columnTruncateLength || this.fftSize, this.audioTrackConstraints = e.audioTrackConstraints, this.smoothingTimeConstant = e.smoothingTimeConstant || 0, this.includeSpectrogram = e.includeSpectrogram !== false, this.includeWaveform = e.includeWaveform === true, !this.includeSpectrogram && !this.includeWaveform)
      throw new Error("Both includeSpectrogram and includeWaveform are false. At least one type of data should be returned.");
  }
  summary() {
    return "microphone";
  }
  static async create(e = {}) {
    if (!K().get("IS_BROWSER"))
      throw new Error("microphone API is only supported in browser environment.");
    let t = new V0(e);
    return await t.start(), t;
  }
  async start() {
    try {
      this.stream = await navigator.mediaDevices.getUserMedia({ audio: this.audioTrackConstraints == null ? true : this.audioTrackConstraints, video: false });
    } catch (n) {
      throw new Error(`Error thrown while initializing video stream: ${n.message}`);
    }
    if (!this.stream)
      throw new Error("Could not obtain audio from microphone.");
    let e = window.AudioContext || window.webkitAudioContext;
    if (this.audioContext = new e(), !this.sampleRateHz)
      this.sampleRateHz = this.audioContext.sampleRate;
    else if (this.audioContext.sampleRate !== this.sampleRateHz)
      throw new Error(`Mismatch in sampling rate: Expected: ${this.sampleRateHz}; Actual: ${this.audioContext.sampleRate}`);
    let t = this.audioContext.createMediaStreamSource(this.stream);
    this.analyser = this.audioContext.createAnalyser(), this.analyser.fftSize = this.fftSize * 2, this.analyser.smoothingTimeConstant = this.smoothingTimeConstant, t.connect(this.analyser), this.freqData = new Float32Array(this.fftSize), this.timeData = new Float32Array(this.fftSize);
  }
  async next() {
    if (this.isClosed)
      return { value: null, done: true };
    let e, t, n = await this.getAudioData();
    if (this.includeSpectrogram) {
      let s = this.flattenQueue(n.freqDataQueue);
      e = this.getTensorFromAudioDataArray(s, [this.numFrames, this.columnTruncateLength, 1]);
    }
    if (this.includeWaveform) {
      let s = this.flattenQueue(n.timeDataQueue);
      t = this.getTensorFromAudioDataArray(s, [this.numFrames * this.fftSize, 1]);
    }
    return { value: { spectrogram: e, waveform: t }, done: false };
  }
  async capture() {
    return (await this.next()).value;
  }
  async getAudioData() {
    let e = [], t = [], n = 0;
    return new Promise((s) => {
      let r = setInterval(() => {
        this.includeSpectrogram && (this.analyser.getFloatFrequencyData(this.freqData), this.freqData[0] === -1 / 0 && s({ freqDataQueue: e, timeDataQueue: t }), e.push(this.freqData.slice(0, this.columnTruncateLength))), this.includeWaveform && (this.analyser.getFloatTimeDomainData(this.timeData), t.push(this.timeData.slice())), ++n === this.numFrames && (clearInterval(r), s({ freqDataQueue: e, timeDataQueue: t }));
      }, this.fftSize / this.sampleRateHz * 1e3);
    });
  }
  stop() {
    this.isClosed || (this.isClosed = true, this.analyser.disconnect(), this.audioContext.close(), this.stream != null && this.stream.getTracks().length > 0 && this.stream.getTracks()[0].stop());
  }
  toArray() {
    throw new Error("Can not convert infinite audio stream to array.");
  }
  getSampleRate() {
    return this.sampleRateHz;
  }
  flattenQueue(e) {
    let t = e[0].length, n = new Float32Array(e.length * t);
    return e.forEach((s, r) => n.set(s, r * t)), n;
  }
  getTensorFromAudioDataArray(e, t) {
    let n = new Float32Array(w.sizeFromShape(t));
    return n.set(e, n.length - e.length), ms(n, t);
  }
};
var W0 = class extends Gt {
  constructor(e, t) {
    if (super(), this.webcamVideoElement = e, this.webcamConfig = t, this.isClosed = true, this.resize = false, this.needToResize())
      if (this.resize = true, this.cropSize = [this.webcamConfig.resizeHeight, this.webcamConfig.resizeWidth], this.cropBoxInd = Zt([0], "int32"), this.webcamConfig.centerCrop) {
        let n = this.webcamConfig.resizeWidth * 1 / this.webcamVideoElement.width, s = this.webcamConfig.resizeHeight * 1 / this.webcamVideoElement.height, r = (1 - n) / 2, a = (1 - s) / 2, i = r + n, o = s + a;
        this.cropBox = Zi([a, r, o, i], [1, 4]);
      } else
        this.cropBox = Zi([0, 0, 1, 1], [1, 4]);
  }
  summary() {
    return "webcam";
  }
  static async create(e, t = {}) {
    if (!K().get("IS_BROWSER"))
      throw new Error("tf.data.webcam is only supported in browser environment.");
    if (!e) {
      if (e = document.createElement("video"), !t.resizeWidth || !t.resizeHeight)
        throw new Error("Please provide webcam video element, or resizeWidth and resizeHeight to create a hidden video element.");
      e.width = t.resizeWidth, e.height = t.resizeHeight;
    }
    let n = new W0(e, t);
    return await n.start(), n;
  }
  async start() {
    this.webcamConfig.facingMode && w.assert(this.webcamConfig.facingMode === "user" || this.webcamConfig.facingMode === "environment", () => `Invalid webcam facing mode: ${this.webcamConfig.facingMode}. Please provide 'user' or 'environment'`);
    try {
      this.stream = await navigator.mediaDevices.getUserMedia({ video: { deviceId: this.webcamConfig.deviceId, facingMode: this.webcamConfig.facingMode ? this.webcamConfig.facingMode : "user", width: this.webcamVideoElement.width, height: this.webcamVideoElement.height } });
    } catch (e) {
      throw e.message = `Error thrown while initializing video stream: ${e.message}`, e;
    }
    if (!this.stream)
      throw new Error("Could not obtain video from webcam.");
    try {
      this.webcamVideoElement.srcObject = this.stream;
    } catch (e) {
      console.log(e), this.webcamVideoElement.src = window.URL.createObjectURL(this.stream);
    }
    return this.webcamVideoElement.play(), this.isClosed = false, new Promise((e) => {
      this.webcamVideoElement.onloadedmetadata = () => {
        e();
      };
    });
  }
  async next() {
    if (this.isClosed)
      return { value: null, done: true };
    let e;
    try {
      e = Pk.fromPixels(this.webcamVideoElement);
    } catch (t) {
      throw new Error(`Error thrown converting video to pixels: ${JSON.stringify(t)}`);
    }
    if (this.resize)
      try {
        return { value: this.cropAndResizeFrame(e), done: false };
      } catch (t) {
        throw new Error(`Error thrown cropping the video: ${t.message}`);
      } finally {
        e.dispose();
      }
    else
      return { value: e, done: false };
  }
  needToResize() {
    return !!(this.webcamConfig.resizeWidth && this.webcamConfig.resizeHeight && (this.webcamVideoElement.width !== this.webcamConfig.resizeWidth || this.webcamVideoElement.height !== this.webcamConfig.resizeHeight));
  }
  cropAndResizeFrame(e) {
    return j(() => {
      let t = Pn(le(e, "float32"), 0), n;
      n = jn.cropAndResize(t, this.cropBox, this.cropBoxInd, this.cropSize, "bilinear");
      let s = n.shape;
      return U(n, s.slice(1));
    });
  }
  async capture() {
    return (await this.next()).value;
  }
  stop() {
    this.stream.getTracks().forEach((t) => t.stop());
    try {
      this.webcamVideoElement.srcObject = null;
    } catch (t) {
      console.log(t), this.webcamVideoElement.src = null;
    }
    this.isClosed = true;
  }
  toArray() {
    throw new Error("Can not convert infinite video stream to array.");
  }
};
var U0 = class {
};
var G0 = class extends Gt {
  split(e) {
    return new kU(this, e);
  }
};
var kU = class extends G0 {
  constructor(e, t) {
    super(), this.upstream = e, this.impl = new SU(e, t);
  }
  summary() {
    return this.impl.summary();
  }
  async next() {
    return this.impl.next();
  }
};
var SU = class extends av {
  constructor(e, t) {
    super(), this.upstream = e, this.separator = t, this.carryover = "";
  }
  summary() {
    return `${this.upstream.summary()} -> Split('${this.separator}')`;
  }
  async pump() {
    let e = await this.upstream.next();
    if (e.done)
      return this.carryover === "" ? false : (this.outputQueue.push(this.carryover), this.carryover = "", true);
    let t = e.value.split(this.separator);
    t[0] = this.carryover + t[0];
    for (let n of t.slice(0, -1))
      this.outputQueue.push(n);
    return this.carryover = t[t.length - 1], true;
  }
};
var IU = class extends Gt {
  decodeUTF8() {
    return new CU(this);
  }
};
var CU = class extends G0 {
  constructor(e) {
    super(), this.upstream = e, this.impl = new NU(e);
  }
  summary() {
    return this.impl.summary();
  }
  async next() {
    return this.impl.next();
  }
};
var NU = class extends av {
  constructor(e) {
    if (super(), this.upstream = e, K().get("IS_BROWSER"))
      this.decoder = new TextDecoder("utf-8");
    else {
      let { StringDecoder: t } = Zw();
      this.decoder = new t("utf8");
    }
  }
  summary() {
    return `${this.upstream.summary()} -> Utf8`;
  }
  async pump() {
    let e = await this.upstream.next(), t;
    if (e.done)
      return false;
    t = e.value;
    let n;
    return K().get("IS_BROWSER") ? n = this.decoder.decode(t, { stream: true }) : n = this.decoder.write(Buffer.from(t.buffer)), this.outputQueue.push(n), true;
  }
};
var H0 = class extends IU {
  constructor(e, t = {}) {
    super(), this.file = e, this.options = t, w.assert(e instanceof Uint8Array || (K().get("IS_BROWSER") ? e instanceof File || e instanceof Blob : false), () => "FileChunkIterator only supports File, Blob and Uint8Array right now."), this.offset = t.offset || 0, this.chunkSize = t.chunkSize || 1024 * 1024;
  }
  summary() {
    return `FileChunks ${this.file}`;
  }
  async next() {
    return this.offset >= (this.file instanceof Uint8Array ? this.file.byteLength : this.file.size) ? { value: null, done: true } : { value: await new Promise((t, n) => {
      let s = this.offset + this.chunkSize;
      if (this.file instanceof Uint8Array)
        t(new Uint8Array(this.file.slice(this.offset, s)));
      else {
        let r = new FileReader();
        r.onload = (i) => {
          let o = r.result;
          if (o instanceof ArrayBuffer && (o = new Uint8Array(o)), !(o instanceof Uint8Array))
            return n(new TypeError("FileReader returned unknown type."));
          t(o);
        }, r.onabort = (i) => n(new Error("Aborted")), r.onerror = (i) => n(new Error(i.type));
        let a = this.file.slice(this.offset, s);
        r.readAsArrayBuffer(a);
      }
      this.offset = s;
    }), done: false };
  }
};
async function TU(e, t = {}, n) {
  let s, r;
  typeof e == "string" ? s = e : (s = e.url, r = $U(e));
  let a = await (n || w.fetch)(s, r);
  if (a.ok) {
    let i = new Uint8Array(await a.arrayBuffer());
    return new H0(i, t);
  } else
    throw new Error(a.statusText);
}
var $U = (e) => ({ method: e.method, headers: e.headers, body: e.body, mode: e.mode, credentials: e.credentials, cache: e.cache, redirect: e.redirect, referrer: e.referrer, integrity: e.integrity });
function q0(e) {
  return typeof e == "string" && e.slice(0, 7) === "file://";
}
var j0 = class extends U0 {
  constructor(e, t = {}) {
    super(), this.input = e, this.options = t;
  }
  async iterator() {
    if (q0(this.input) && K().get("IS_NODE")) {
      let e = og();
      this.input = e.readFileSync(this.input.slice(7));
    }
    return new H0(this.input, this.options);
  }
};
var K0 = class extends U0 {
  constructor(e, t = {}) {
    super(), this.url = e, this.fileOptions = t;
  }
  async iterator() {
    return q0(this.url) ? new j0(this.url, this.fileOptions).iterator() : TU(this.url, this.fileOptions);
  }
};
function _U(e, t = {}) {
  return new B0(new K0(e), t);
}
function AU(e) {
  let t = rv(e);
  return $n(async () => t);
}
function EU(e) {
  return $n(async () => {
    let t = await e();
    return rv(() => t.next());
  });
}
async function RU(e, t) {
  return W0.create(e, t);
}
async function DU(e) {
  return V0.create(e);
}
var FU = "0.0.0";
function be(e, t) {
  Array.isArray(e) || (e = [e]), e.forEach((n) => {
    n != null && w.assert(n.dtype !== "complex64", () => `${t} does not support complex64 tensors in the CPU backend.`);
  });
}
var OU = ws.whereImpl;
var X0 = class extends il {
  constructor() {
    super(), this.blockSize = 48, this.firstUse = true, this.data = new Yd(this, ds());
  }
  nextDataId() {
    return X0.nextDataId++;
  }
  write(e, t, n) {
    this.firstUse && (this.firstUse = false, K().get("IS_NODE") && C.warn(`
============================
Hi, looks like you are running TensorFlow.js in Node.js. To speed things up dramatically, install our node backend, visit https://github.com/tensorflow/tfjs-node for more details. 
============================`));
    let s = { id: this.nextDataId() };
    return this.data.set(s, { values: e, dtype: n, refCount: 1 }), s;
  }
  makeTensorInfo(e, t, n) {
    let s;
    if (t === "string" && n != null && n.length > 0 && w.isString(n[0])) {
      let r = n.map((a) => w.encodeString(a));
      s = this.write(r, e, t);
    } else
      s = this.write(n, e, t);
    return { dataId: s, shape: e, dtype: t };
  }
  refCount(e) {
    return this.data.has(e) ? this.data.get(e).refCount : 0;
  }
  incRef(e) {
    let t = this.data.get(e);
    t.refCount++;
  }
  decRef(e) {
    if (this.data.has(e)) {
      let t = this.data.get(e);
      t.refCount--;
    }
  }
  move(e, t, n, s, r) {
    this.data.set(e, { values: t, dtype: s, refCount: r });
  }
  numDataIds() {
    return this.data.numDataIds();
  }
  async read(e) {
    return this.readSync(e);
  }
  readSync(e) {
    let { dtype: t, complexTensorInfos: n } = this.data.get(e);
    if (t === "complex64") {
      let s = this.readSync(n.real.dataId), r = this.readSync(n.imag.dataId);
      return C.mergeRealAndImagArrays(s, r);
    }
    return this.data.get(e).values;
  }
  bufferSync(e) {
    let t = this.readSync(e.dataId);
    if (e.dtype === "string")
      try {
        let n = t.map((s) => w.decodeString(s));
        return Ae(e.shape, e.dtype, n);
      } catch (n) {
        throw new Error("Failed to decode encoded string bytes into utf-8");
      }
    return Ae(e.shape, e.dtype, t);
  }
  makeOutput(e, t, n) {
    return ds().makeTensorFromTensorInfo(this.makeTensorInfo(t, n, e), this);
  }
  disposeData(e, t = false) {
    if (this.data.has(e)) {
      if (this.data.get(e).refCount--, !t && this.data.get(e).refCount > 0)
        return false;
      let { complexTensorInfos: n } = this.data.get(e);
      n != null && (this.disposeData(n.real.dataId, true), this.disposeData(n.imag.dataId, true)), this.data.delete(e);
    }
    return true;
  }
  disposeIntermediateTensorInfo(e) {
    this.disposeData(e.dataId);
  }
  async time(e) {
    let t = w.now();
    return e(), { kernelMs: w.now() - t };
  }
  memory() {
    return { unreliable: true, reasons: ["The reported memory is an upper bound. Due to automatic garbage collection, the true allocated memory may be less."] };
  }
  where(e) {
    be([e], "where");
    let t = this.readSync(e.dataId);
    return OU(e.shape, t);
  }
  dispose() {
  }
  floatPrecision() {
    return 32;
  }
  epsilon() {
    return super.epsilon();
  }
};
var Y0 = X0;
Y0.nextDataId = 0;
var iv = {};
Ee(iv, { addImpl: () => Z0, bincountImpl: () => uv, bincountReduceImpl: () => J0, ceilImpl: () => eC, concatImpl: () => lv, equalImpl: () => tC, expImpl: () => sC, expm1Impl: () => aC, floorImpl: () => iC, gatherNdImpl: () => oC, gatherV2Impl: () => uC, greaterEqualImpl: () => cC, greaterImpl: () => lC, lessEqualImpl: () => pC, lessImpl: () => dC, linSpaceImpl: () => hC, logImpl: () => fC, maxImpl: () => mC, maximumImpl: () => gC, minimumImpl: () => bC, multiplyImpl: () => cv, negImpl: () => yC, notEqualImpl: () => vC, prodImpl: () => xC, rangeImpl: () => pv, rsqrtImpl: () => wC, scatterImpl: () => Ki, sigmoidImpl: () => SG, simpleAbsImpl: () => Q0, sliceImpl: () => Bd, sparseFillEmptyRowsImpl: () => SC, sparseReshapeImpl: () => IC, sparseSegmentReductionImpl: () => hv, sqrtImpl: () => NG, squaredDifferenceImpl: () => CC, stridedSliceImpl: () => NC, stringNGramsImpl: () => TC, stringSplitImpl: () => $C, stringToHashBucketFastImpl: () => _C, subImpl: () => AC, tileImpl: () => EC, topKImpl: () => DC, transposeImpl: () => dv, uniqueImpl: () => FC });
function Q0(e) {
  let t = new Float32Array(e.length);
  for (let n = 0; n < e.length; ++n)
    t[n] = Math.abs(e[n]);
  return t;
}
var PU = (e) => {
  let { x: t } = e.inputs, n = e.backend;
  be(t, "abs");
  let s = new Float32Array(w.sizeFromShape(t.shape)), r = n.data.get(t.dataId).values;
  return s = Q0(r), n.makeOutput(s, t.shape, t.dtype);
};
var zU = { kernelName: po, backendName: "cpu", kernelFunc: PU };
function Et(e) {
  return (t, n, s, r, a) => {
    let i = C.assertAndGetBroadcastShape(t, n), o = i.length, u = w.computeStrides(i), l = w.sizeFromShape(i), c = w.getTypedArrayFromDType(a, l), p = t.length, d = n.length, h = w.computeStrides(t), f = w.computeStrides(n), m = C.getBroadcastDims(t, i), g = C.getBroadcastDims(n, i);
    if (m.length + g.length === 0)
      for (let b = 0; b < c.length; ++b)
        c[b] = e(s[b % s.length], r[b % r.length]);
    else
      for (let b = 0; b < c.length; ++b) {
        let y = w.indexToLoc(b, o, u), v = y.slice(-p);
        m.forEach(($) => v[$] = 0);
        let x = w.locToIndex(v, p, h), k = y.slice(-d);
        g.forEach(($) => k[$] = 0);
        let I = w.locToIndex(k, d, f);
        c[b] = e(s[x], r[I]);
      }
    return [c, i];
  };
}
function En(e) {
  let { inputs: t, backend: n } = e, { real: s, imag: r } = t, a = n.data.get(s.dataId).values, i = n.data.get(r.dataId).values, o = n.makeTensorInfo(s.shape, "complex64"), u = n.data.get(o.dataId);
  return u.complexTensorInfos = { real: n.makeTensorInfo(s.shape, "float32", a), imag: n.makeTensorInfo(r.shape, "float32", i) }, o;
}
var MU = { kernelName: ep, backendName: "cpu", kernelFunc: En };
function Ld(e, t, n = "float32") {
  if (n === "complex64") {
    let r = Ld(e, t, "float32"), a = Ld(e, t, "float32");
    return En({ inputs: { real: r, imag: a }, backend: e });
  }
  let s = w.makeZerosTypedArray(w.sizeFromShape(t), n);
  return e.makeTensorInfo(t, n, s);
}
function Os(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  return n.incRef(s.dataId), { dataId: s.dataId, shape: s.shape, dtype: s.dtype };
}
var LU = { kernelName: Wa, backendName: "cpu", kernelFunc: Os };
function ga(e) {
  let { inputs: t, backend: n } = e, { input: s } = t, r = n.data.get(s.dataId).complexTensorInfos.real, a = n.data.get(r.dataId).values;
  return n.makeTensorInfo(r.shape, r.dtype, a);
}
var BU = { kernelName: lp, backendName: "cpu", kernelFunc: ga };
function xr(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { dtype: a } = s;
  if (a === "complex64") {
    if (r.dtype === "complex64")
      return Os({ inputs: { x: r }, backend: n });
    let i = Ld(n, r.shape, r.dtype), o = xr({ inputs: { x: r }, backend: n, attrs: { dtype: "float32" } }), u = En({ inputs: { real: o, imag: i }, backend: n });
    return n.disposeIntermediateTensorInfo(i), n.disposeIntermediateTensorInfo(o), u;
  }
  if (r.dtype === "complex64") {
    let i = ga({ inputs: { input: r }, backend: n }), o = xr({ inputs: { x: i }, backend: n, attrs: { dtype: a } });
    return n.disposeIntermediateTensorInfo(i), o;
  }
  if (!w.hasEncodingLoss(r.dtype, a)) {
    let i = Os({ inputs: { x: r }, backend: n });
    return { dataId: i.dataId, shape: i.shape, dtype: a };
  }
  if (a === "int32") {
    let i = n.data.get(r.dataId).values, o = Int32Array.from(i);
    return n.makeTensorInfo(r.shape, "int32", o);
  }
  if (a === "bool") {
    let i = n.data.get(r.dataId).values, o = w.toTypedArray([0], r.dtype), [u, l] = Et((c, p) => c !== p ? 1 : 0)(r.shape, [], i, o, "bool");
    return n.makeTensorInfo(l, "bool", u);
  }
  throw new Error(`Error in Cast: failed to cast ${r.dtype} to ${a}`);
}
var VU = { kernelName: Ta, backendName: "cpu", kernelFunc: xr };
function Ht(e, t, n, s) {
  return n == null ? ({ inputs: r, backend: a }) => {
    let { a: i, b: o } = r, u = a;
    be([i, o], e);
    let l = u.data.get(i.dataId).values, c = u.data.get(o.dataId).values, p = i.dtype === "string" ? C.fromUint8ToStringArray(l) : l, d = i.dtype === "string" ? C.fromUint8ToStringArray(c) : c, h = s || i.dtype, [f, m] = t(i.shape, o.shape, p, d, h);
    return u.makeTensorInfo(m, h, f);
  } : ({ inputs: r, backend: a }) => {
    let { a: i, b: o } = r, u = a;
    if (i.dtype === "complex64" || o.dtype === "complex64") {
      let l = xr({ inputs: { x: i }, backend: u, attrs: { dtype: "complex64" } }), c = u.data.get(l.dataId), p = c.complexTensorInfos.real, d = c.complexTensorInfos.imag, h = u.data.get(p.dataId).values, f = u.data.get(d.dataId).values, m = xr({ inputs: { x: o }, backend: u, attrs: { dtype: "complex64" } }), g = u.data.get(m.dataId), b = g.complexTensorInfos.real, y = g.complexTensorInfos.imag, v = u.data.get(b.dataId).values, x = u.data.get(y.dataId).values, [k, I, $] = n(i.shape, o.shape, h, f, v, x), R = u.makeTensorInfo($, "float32", k), E = u.makeTensorInfo($, "float32", I), P = En({ inputs: { real: R, imag: E }, backend: u });
      return u.disposeIntermediateTensorInfo(l), u.disposeIntermediateTensorInfo(m), u.disposeIntermediateTensorInfo(R), u.disposeIntermediateTensorInfo(E), P;
    } else {
      let l = u.data.get(i.dataId).values, c = u.data.get(o.dataId).values, p = s || i.dtype, [d, h] = t(i.shape, o.shape, l, c, p);
      return u.makeTensorInfo(h, p, d);
    }
  };
}
function ov(e) {
  return (t, n, s, r, a, i) => {
    let o = C.assertAndGetBroadcastShape(t, n), u = w.sizeFromShape(o), l = o.length, c = w.computeStrides(o), p = w.getTypedArrayFromDType("float32", u), d = w.getTypedArrayFromDType("float32", u), h = C.getBroadcastDims(t, o), f = C.getBroadcastDims(n, o), m = C.mergeRealAndImagArrays(s, r), g = C.mergeRealAndImagArrays(a, i), b = t.length, y = w.computeStrides(t), v = n.length, x = w.computeStrides(n);
    if (h.length + f.length === 0)
      for (let k = 0; k < p.length; k++) {
        let I = k % m.length, $ = k % g.length, R = e(m[I * 2], m[I * 2 + 1], g[$ * 2], g[$ * 2 + 1]);
        p[k] = R.real, d[k] = R.imag;
      }
    else
      for (let k = 0; k < p.length; k++) {
        let I = w.indexToLoc(k, l, c), $ = I.slice(-b);
        h.forEach((O) => $[O] = 0);
        let R = w.locToIndex($, b, y), E = I.slice(-v);
        f.forEach((O) => E[O] = 0);
        let P = w.locToIndex(E, v, x), A = e(m[R * 2], m[R * 2 + 1], g[P * 2], g[P * 2 + 1]);
        p[k] = A.real, d[k] = A.imag;
      }
    return [p, d, o];
  };
}
var Z0 = Et((e, t) => e + t);
var WU = ov((e, t, n, s) => ({ real: e + n, imag: t + s }));
var io = Ht(Sr, Z0, WU);
var UU = { kernelName: Sr, backendName: "cpu", kernelFunc: io };
function uv(e, t, n, s, r) {
  let a = w.sizeFromShape(s), i = w.makeZerosTypedArray(r, n);
  for (let o = 0; o < e.length; o++) {
    let u = e[o];
    if (u < 0)
      throw new Error("Input x must be non-negative!");
    u >= r || (a > 0 ? i[u] += t[o] : i[u] += 1);
  }
  return i;
}
function J0(e, t, n, s = false) {
  let r = e.shape[0], a = e.shape[1], i = Ae([r, n], t.dtype);
  for (let o = 0; o < r; o++)
    for (let u = 0; u < a; u++) {
      let l = e.get(o, u);
      if (l < 0)
        throw new Error("Input x must be non-negative!");
      l >= n || (s ? i.set(1, o, l) : t.size > 0 ? i.set(i.get(o, l) + t.get(o, u), o, l) : i.set(i.get(o, l) + 1, o, l));
    }
  return i;
}
function Er(e) {
  return (t, n, s) => {
    let r = w.getTypedArrayFromDType(n, t.length);
    for (let a = 0; a < t.length; ++a)
      r[a] = e(t[a], s);
    return r;
  };
}
function st(e, t, n) {
  return ({ inputs: s, attrs: r, backend: a }) => {
    let { x: i } = s;
    if (be(i, e), i.dtype === "string" || n === "string")
      throw new Error("unaryKernelFunc does not support string input/output");
    let o = a, u = o.data.get(i.dataId).values, l = w.sizeFromShape(i.shape), c = n || i.dtype, p = w.getArrayFromDType(c, l);
    for (let d = 0; d < l; ++d)
      p[d] = t(u[d], r);
    return o.makeTensorInfo(i.shape, c, p);
  };
}
function ru(e, t, n) {
  return ({ inputs: s, attrs: r, backend: a }) => {
    let { x: i } = s;
    if (be(i, e), i.dtype === "string" || n === "string")
      throw new Error("unaryKernelFunc does not support string input/output");
    let o = a, u = o.data.get(i.dataId).values, l = n || i.dtype, c = t(u, l, r);
    return o.makeTensorInfo(i.shape, l, c);
  };
}
var eC = Er((e) => Math.ceil(e));
var GU = ru($a, eC);
var HU = { kernelName: $a, backendName: "cpu", kernelFunc: GU };
function lv(e, t, n, s) {
  let r = w.getArrayFromDType(n, w.sizeFromShape(t));
  if (s && n !== "string") {
    let a = 0;
    e.forEach((i) => {
      let o = w.sizeFromShape(i.shape);
      r.set(i.vals, a), a += o;
    });
  } else {
    let a = 0;
    e.forEach((i) => {
      let o = n === "string" ? C.fromUint8ToStringArray(i.vals) : i.vals, u = 0;
      for (let l = 0; l < i.shape[0]; ++l) {
        let c = l * t[1] + a;
        for (let p = 0; p < i.shape[1]; ++p)
          r[c + p] = o[u++];
      }
      a += i.shape[1];
    });
  }
  return r;
}
var tC = Et((e, t) => e === t ? 1 : 0);
var nC = Ht(yo, tC, null, "bool");
var qU = { kernelName: yo, backendName: "cpu", kernelFunc: nC };
var sC = Er((e) => Math.exp(e));
var rC = ru(za, sC, "float32");
var jU = { kernelName: za, backendName: "cpu", kernelFunc: rC };
var aC = Er((e) => Math.expm1(e));
var KU = ru(xo, aC);
var XU = { kernelName: xo, backendName: "cpu", kernelFunc: KU };
var iC = Er((e) => Math.floor(e));
var YU = ru(Ma, iC);
var QU = { kernelName: Ma, backendName: "cpu", kernelFunc: YU };
function oC(e, t, n, s, r, a, i, o, u) {
  let l = Ae([s, a], n);
  for (let c = 0; c < s; c++) {
    let p = [], d = 0;
    for (let h = 0; h < r; h++) {
      let f = e[c * r + h];
      d += f * i[h], p.push(f);
    }
    if (d < 0 || d >= u / a)
      throw new Error(`Invalid indices: ${p} does not index into ${o}`);
    for (let h = 0; h < a; h++)
      l.values[c * a + h] = t.get(...t.indexToLoc(d * a + h));
  }
  return l;
}
function uC(e, t, n) {
  let s = Ae(n, e.dtype);
  for (let r = 0; r < s.size; ++r) {
    let i = s.indexToLoc(r).slice(), o = i[0], u = i[2], l = t.locToIndex([o, u]);
    i[2] = t.values[l];
    let c = e.locToIndex(i);
    0 <= c && c < e.values.length && (s.values[r] = e.values[c]);
  }
  return s;
}
var lC = Et((e, t) => e > t ? 1 : 0);
var ZU = Ht(Io, lC, null, "bool");
var JU = { kernelName: Io, backendName: "cpu", kernelFunc: ZU };
var cC = Et((e, t) => e >= t ? 1 : 0);
var eG = Ht(Va, cC, null, "bool");
var tG = { kernelName: Va, backendName: "cpu", kernelFunc: eG };
var dC = Et((e, t) => e < t ? 1 : 0);
var nG = Ht(Co, dC, null, "bool");
var sG = { kernelName: Co, backendName: "cpu", kernelFunc: nG };
var pC = Et((e, t) => e <= t ? 1 : 0);
var rG = Ht(No, pC, null, "bool");
var aG = { kernelName: No, backendName: "cpu", kernelFunc: rG };
function hC(e, t, n) {
  let s = (t - e) / (n - 1), r = w.makeZerosTypedArray(n, "float32");
  r[0] = e;
  for (let a = 1; a < r.length; a++)
    r[a] = r[a - 1] + s;
  return r;
}
var fC = Er((e) => Math.log(e));
var iG = ru(Ga, fC);
var oG = { kernelName: Ga, backendName: "cpu", kernelFunc: iG };
function mC(e, t, n, s) {
  let r = w.getTypedArrayFromDType(s, w.sizeFromShape(n));
  for (let a = 0; a < r.length; ++a) {
    let i = a * t, o = e[i];
    for (let u = 0; u < t; ++u) {
      let l = e[i + u];
      (Number.isNaN(l) || l > o) && (o = l);
    }
    r[a] = o;
  }
  return r;
}
var gC = Et((e, t) => Math.max(e, t));
var uG = Ht(qa, gC);
var lG = { kernelName: qa, backendName: "cpu", kernelFunc: uG };
var bC = Et((e, t) => Math.min(e, t));
var cG = Ht(Ya, bC);
var dG = { kernelName: Ya, backendName: "cpu", kernelFunc: cG };
var cv = Et((e, t) => e * t);
var pG = ov((e, t, n, s) => ({ real: e * n - t * s, imag: e * s + t * n }));
var Zp = Ht(Za, cv, pG);
var hG = { kernelName: Za, backendName: "cpu", kernelFunc: Zp };
function yC(e, t, n) {
  let s = w.createScalarValue(-1, n);
  return cv([], t, s, e, n);
}
function fG(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  be(s, "neg");
  let r = n.data.get(s.dataId).values, [a, i] = yC(r, s.shape, s.dtype);
  return n.makeTensorInfo(i, s.dtype, a);
}
var mG = { kernelName: $o, backendName: "cpu", kernelFunc: fG };
var vC = Et((e, t) => e !== t ? 1 : 0);
var gG = Ht(_o, vC, null, "bool");
var bG = { kernelName: _o, backendName: "cpu", kernelFunc: gG };
function dv(e, t, n, s, r) {
  let a = t.length, i = w.sizeFromShape(t), o = w.computeStrides(t), u = w.computeStrides(r), l = w.getTypedArrayFromDType(n, w.sizeFromShape(r));
  for (let c = 0; c < i; ++c) {
    let p = w.indexToLoc(c, a, o), d = new Array(p.length);
    for (let f = 0; f < d.length; f++)
      d[f] = p[s[f]];
    let h = w.locToIndex(d, a, u);
    l[h] = e[c];
  }
  return l;
}
function wn(e) {
  let { inputs: t, attrs: n, backend: s } = e, { x: r } = t, { perm: a } = n;
  be(r, "transpose");
  let i = r.shape.length, o = new Array(i);
  for (let p = 0; p < o.length; p++)
    o[p] = r.shape[a[p]];
  let u = s.data.get(r.dataId).values, l = dv(u, r.shape, r.dtype, a, o);
  return { dataId: s.write(l, o, r.dtype), shape: o, dtype: r.dtype };
}
var yG = { kernelName: mi, backendName: "cpu", kernelFunc: wn };
function xC(e, t, n, s) {
  let [r, a] = C.computeOutAndReduceShapes(e, s), i = cn(t, "int32"), o = w.makeZerosTypedArray(w.sizeFromShape(r), i), u = w.sizeFromShape(a);
  for (let l = 0; l < o.length; ++l) {
    let c = l * u, p = 1;
    for (let d = 0; d < u; ++d)
      p *= n[c + d];
    o[l] = p;
  }
  return { outVals: o, outShape: r, outDtype: i };
}
function vG(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  be(r, "prod");
  let o = r.shape.length, u = w.parseAxisParam(a, r.shape), l = C.getAxesPermutation(u, o), c = u, p = r, d = [];
  l != null && (p = wn({ inputs: { x: r }, backend: n, attrs: { perm: l } }), d.push(p), c = C.getInnerMostAxes(c.length, o));
  let h = n.data.get(p.dataId).values, { outVals: f, outShape: m, outDtype: g } = xC(p.shape, p.dtype, h, c), b = m;
  return i && (b = C.expandShapeToKeepDim(m, u)), d.forEach((y) => n.disposeIntermediateTensorInfo(y)), n.makeTensorInfo(b, g, f);
}
var xG = { kernelName: ni, backendName: "cpu", kernelFunc: vG };
function pv(e, t, n, s) {
  let r = e === t, a = e < t && n < 0, i = t < e && n > 1;
  if (r || a || i)
    return w.makeZerosTypedArray(0, s);
  let o = Math.abs(Math.ceil((t - e) / n)), u = w.makeZerosTypedArray(o, s);
  t < e && n === 1 && (n = -1), u[0] = e;
  for (let l = 1; l < u.length; l++)
    u[l] = u[l - 1] + n;
  return u;
}
var wC = Er((e) => 1 / Math.sqrt(e));
var wG = ru(ii, wC);
var kG = { kernelName: ii, backendName: "cpu", kernelFunc: wG };
function Ki(e, t, n, s, r, a, i, o, u, l) {
  let c = [s / r, r], p = e.values, d = t.values;
  if (s === 0)
    return Ae(n, t.dtype);
  let h = Ae(c, t.dtype);
  typeof u == "string" || typeof u == "number" ? h.values.fill(u) : typeof u == "boolean" && h.values.fill(+u);
  for (let f = 0; f < a; f++) {
    let m = [], g = 0;
    for (let b = 0; b < i; b++) {
      let y = p[f * i + b];
      m.push(y), g += y * o[b];
    }
    if (g < 0 || g >= s / r)
      throw new Error(`Invalid indices: ${m} does not index into ${n}`);
    for (let b = 0; b < r; b++)
      l ? h.values[g * r + b] += d[f * r + b] : h.values[g * r + b] = t.rank === 0 ? d[0] : d[f * r + b];
  }
  return h;
}
var SG = Er((e) => 1 / (1 + Math.exp(-e)));
var kC = st(ui, (e) => 1 / (1 + Math.exp(-e)));
var IG = { kernelName: ui, backendName: "cpu", kernelFunc: kC };
function Bd(e, t, n, s, r) {
  let a = wt.isSliceContinous(s, t, n), i = w.sizeFromShape(n), o = w.computeStrides(s);
  if (a) {
    let p = wt.computeFlatOffset(t, o);
    return r === "string" ? e.slice(p, p + i) : e.subarray(p, p + i);
  }
  let u = r === "string" ? C.fromUint8ToStringArray(e) : e, l = Ae(s, r, u), c = Ae(n, r);
  for (let p = 0; p < c.size; ++p) {
    let d = c.indexToLoc(p), h = d.map((f, m) => f + t[m]);
    c.set(l.get(...h), ...d);
  }
  return r === "string" ? C.fromStringArrayToUint8(c.values) : c.values;
}
function ba(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { begin: a, size: i } = s;
  be(r, "slice");
  let [o, u] = wt.parseSliceParams(r, a, i);
  wt.assertParamsValid(r, o, u);
  let l = n.data.get(r.dataId).values, c = Bd(l, o, u, r.shape, r.dtype);
  return n.makeTensorInfo(u, r.dtype, c);
}
var CG = { kernelName: Bo, backendName: "cpu", kernelFunc: ba };
function SC(e, t, n, s, r, a, i) {
  let o = t[0], u = a[0], l = new Array(u), c = new Array(o), p = t[1];
  if (u === 0) {
    if (o !== 0)
      throw new Error(C.getSparseFillEmptyRowsIndicesDenseShapeMismatch(o));
    let g = w.getArrayFromDType(n, 0), b = w.getArrayFromDType(r, 0);
    return [g, [0, p], b, l, c];
  }
  let d = true, h = 0, f = new Array(u).fill(0);
  for (let g = 0; g < o; ++g) {
    let b = e[g * p];
    if (b < 0)
      throw new Error(C.getSparseFillEmptyRowsNegativeIndexErrorMessage(g, b));
    if (b >= u)
      throw new Error(C.getSparseFillEmptyRowsOutOfRangeIndexErrorMessage(g, b, u));
    ++f[b], d = d && b >= h, h = b;
  }
  let m = true;
  for (let g = 0; g < u; ++g) {
    let b = f[g] === 0;
    l[g] = b, m = m && !b, f[g] = Math.max(f[g], 1), g > 0 && (f[g] += f[g - 1]);
  }
  if (m && d) {
    let g = e, b = s;
    for (let y = 0; y < o; ++y)
      c[y] = y;
    return [g, [o, p], b, l, c];
  } else {
    let g = f[u - 1], b = w.getArrayFromDType(n, g * p), y = w.getArrayFromDType(r, g), v = new Array(u).fill(0);
    for (let x = 0; x < o; ++x) {
      let k = e[x * p], I = v[k], $ = (k === 0 ? 0 : f[k - 1]) + I;
      v[k]++;
      for (let R = 0; R < p; ++R)
        b[$ * p + R] = e[x * p + R];
      y[$] = s[x], c[x] = $;
    }
    for (let x = 0; x < u; ++x)
      if (v[x] === 0) {
        let I = x === 0 ? 0 : f[x - 1];
        b[I * p + 0] = x;
        for (let $ = 1; $ < p; ++$)
          b[I * p + $] = 0;
        y[I] = i;
      }
    return [b, [g, p], y, l, c];
  }
}
function IC(e, t, n, s, r) {
  let a = w.sizeFromShape(s), i = t[0], o = r.length, u = [], l = 1, c = -1;
  for (let g = 0; g < o; ++g) {
    let b = r[g];
    if (b === -1) {
      if (c !== -1)
        throw new Error(C.getSparseReshapeMultipleNegativeOneOutputDimErrorMessage(c, g));
      c = g, u.push(1);
    } else {
      if (b < 0)
        throw new Error(C.getSparseReshapeNegativeOutputDimErrorMessage(g, b));
      l *= b, u.push(b);
    }
  }
  if (c !== -1) {
    if (l <= 0)
      throw new Error(C.getSparseReshapeEmptyTensorZeroOutputDimErrorMessage());
    let g = Math.trunc(a / l);
    if (l * g !== a)
      throw new Error(C.getSparseReshapeInputOutputMultipleErrorMessage(s, u));
    u[c] = g;
  }
  if (w.sizeFromShape(u) !== a)
    throw new Error(C.getSparseReshapeInputOutputMismatchErrorMessage(s, u));
  let d = s.length, h = [];
  if (d > 0) {
    h[d - 1] = 1;
    for (let g = d - 2; g >= 0; --g)
      h[g] = h[g + 1] * s[g + 1];
  }
  let f = [];
  if (o > 0) {
    f[o - 1] = 1;
    for (let g = o - 2; g >= 0; --g)
      f[g] = f[g + 1] * u[g + 1];
  }
  let m = w.getArrayFromDType(n, i * o);
  for (let g = 0; g < i; ++g) {
    let b = 0;
    for (let y = 0; y < d; ++y)
      b += e[g * d + y] * h[y];
    for (let y = 0; y < o; ++y)
      m[g * o + y] = Math.trunc(b / f[y]), b %= f[y];
  }
  return [m, [i, o], u];
}
function hv(e, t, n, s, r, a = false, i = 0) {
  let o = s.length, u = [t[0], e.length / t[0]], l = u[1], p = o > 0 ? r[o - 1] + 1 : 0;
  if (p < 0)
    throw new Error(C.getSparseSegmentReductionNegativeSegmentIdsErrorMessage());
  let d = t.slice();
  d[0] = p;
  let h = d.reduce((v, x) => v * x, 1), f = w.getArrayFromDType(n, h);
  if (o === 0)
    return p > 0 && f.fill(i), [f, d];
  if (p <= 0)
    throw new Error(C.getSparseSegmentReductionNegativeSegmentIdsErrorMessage());
  let m = 0, g = 1, b = 0, y = r[m];
  for (; ; ) {
    let v = 0;
    if (g < o) {
      if (v = r[g], y === v) {
        ++g;
        continue;
      }
      if (y >= v)
        throw new Error(C.getSparseSegmentReductionNonIncreasingSegmentIdsErrorMessage());
    }
    if (y < 0 || y >= p)
      throw new Error(C.getSparseSegmentReductionSegmentIdOutOfRangeErrorMessage(y, p));
    y > b && f.fill(i, b * l, y * l);
    for (let x = m; x < g; ++x) {
      let k = s[x];
      if (k < 0 || k >= u[0])
        throw new Error(C.getSparseSegmentReductionIndicesOutOfRangeErrorMessage(x, s[x], u[0]));
      for (let I = 0; I < l; I++)
        f[y * l + I] += e[k * l + I];
    }
    if (a)
      for (let x = 0; x < l; x++)
        f[y * l + x] /= g - m;
    if (m = g, ++g, b = y + 1, y = v, g > o)
      break;
  }
  return b < p && f.fill(i, b * l, p * l), [f, d];
}
var NG = Er((e) => Math.sqrt(e));
var TG = st(li, (e) => Math.sqrt(e));
var $G = { kernelName: li, backendName: "cpu", kernelFunc: TG };
var CC = Et((e, t) => {
  let n = e - t;
  return n * n;
});
var _G = Ht(pi, CC);
var AG = { kernelName: pi, backendName: "cpu", kernelFunc: _G };
function NC(e, t, n, s) {
  let r = Ae(e, t.dtype);
  for (let a = 0; a < r.size; a++) {
    let i = r.indexToLoc(a), o = new Array(i.length);
    for (let u = 0; u < o.length; u++)
      o[u] = i[u] * n[u] + s[u];
    r.set(t.get(...o), ...i);
  }
  return r;
}
var EG = class {
  constructor(e, t, n, s, r, a) {
    this.separator = w.encodeString(e), this.nGramWidths = t, this.leftPad = w.encodeString(n), this.rightPad = w.encodeString(s), this.padWidth = r, this.preserveShort = a;
  }
  getPadWidth(e) {
    return Math.min(this.padWidth < 0 ? e - 1 : this.padWidth, e - 1);
  }
  getNumNGrams(e, t) {
    let n = this.getPadWidth(t);
    return Math.max(0, e + 2 * n - t + 1);
  }
  createNGrams(e, t, n, s, r, a) {
    for (let i = 0; i < r; ++i) {
      let o = this.getPadWidth(a), u = Math.max(0, o - i), l = Math.max(0, o - (r - (i + 1))), c = a - (u + l), p = t + (u > 0 ? 0 : i - o), d = 0;
      d += u * this.leftPad.length;
      for (let b = 0; b < c; ++b)
        d += e[p + b].length;
      d += l * this.rightPad.length, d += (u + l + c - 1) * this.separator.length, n[s + i] = new Uint8Array(d);
      let f = n[s + i], m = 0, g = (b) => b.forEach((y) => f[m++] = y);
      for (let b = 0; b < u; ++b)
        g(this.leftPad), g(this.separator);
      for (let b = 0; b < c - 1; ++b)
        g(e[p + b]), g(this.separator);
      if (c > 0) {
        g(e[p + c - 1]);
        for (let b = 0; b < l; ++b)
          g(this.separator), g(this.rightPad);
      } else {
        for (let b = 0; b < l - 1; ++b)
          g(this.rightPad), g(this.separator);
        g(this.rightPad);
      }
    }
  }
  compute(e, t) {
    let n = e.length, s = t.length;
    if (s > 0) {
      let o = t[0];
      if (o !== 0)
        throw new Error(`First split value must be 0, got ${o}`);
      for (let u = 1; u < s; ++u) {
        let l = t[u] >= o;
        if (l = l && t[u] <= n, !l)
          throw new Error(`Invalid split value ${t[u]}, must be in [${o}, ${n}]`);
        o = t[u];
      }
      if (o !== n)
        throw new Error(`Last split value must be data size. Expected ${n}, got ${o}`);
    }
    let r = s - 1, a = w.getArrayFromDType("int32", s);
    if (n === 0 || s === 0) {
      let o = new Array(n);
      for (let u = 0; u <= r; ++u)
        a[u] = 0;
      return [o, a];
    }
    a[0] = 0;
    for (let o = 1; o <= r; ++o) {
      let u = t[o] - t[o - 1], l = 0;
      this.nGramWidths.forEach((c) => {
        l += this.getNumNGrams(u, c);
      }), this.preserveShort && u > 0 && l === 0 && (l = 1), a[o] = a[o - 1] + l;
    }
    let i = new Array(a[r]);
    for (let o = 0; o < r; ++o) {
      let u = t[o], l = a[o];
      if (this.nGramWidths.forEach((c) => {
        let p = t[o + 1] - t[o], d = this.getNumNGrams(p, c);
        this.createNGrams(e, u, i, l, d, c), l += d;
      }), this.preserveShort && l === a[o]) {
        let c = t[o + 1] - t[o];
        if (c === 0)
          continue;
        let p = c + 2 * this.padWidth, d = 1;
        this.createNGrams(e, u, i, l, d, p);
      }
    }
    return [i, a];
  }
};
function TC(e, t, n, s, r, a, i, o) {
  return new EG(n, s, r, a, i, o).compute(e, t);
}
function RG(e, t, n, s) {
  if (!e.length)
    return;
  if (t.length === 0) {
    for (let a = 0; a < e.length; ++a)
      s.push(e.subarray(a, a + 1));
    return;
  }
  if (t.length === 1) {
    let a = t[0], i = e.indexOf(a);
    for (; i !== -1; ) {
      let o = e.subarray(0, i);
      (!n || o.length !== 0) && s.push(o), e = e.subarray(i + 1), i = e.indexOf(a);
    }
    (!n || e.length !== 0) && s.push(e);
    return;
  }
  let r = 0;
  for (let a = 0; a < e.length + 1; a++)
    if (a === e.length || t.indexOf(e[a]) !== -1) {
      let i = e.subarray(r, a);
      (!n || i.length !== 0) && s.push(i), r = a + 1;
    }
}
function $C(e, t, n) {
  let s = e.length, r = [], a = 0, i = 0, o = new Array(s);
  for (let d = 0; d < s; ++d) {
    let h = r.length;
    RG(e[d], t, n, r);
    let f = r.length - h;
    o[d] = f, a += f, i = Math.max(i, f);
  }
  let u = w.getArrayFromDType("int32", a * 2), l = new Array(a), c = [s, i], p = 0;
  for (let d = 0; d < s; ++d)
    for (let h = 0; h < o[d]; ++h)
      u[p * 2] = d, u[p * 2 + 1] = h, l[p] = r[p], ++p;
  return [u, l, c];
}
function _C(e, t) {
  let n = w.getArrayFromDType("int32", e.length);
  for (let s = 0; s < e.length; ++s)
    n[s] = w.fingerPrint64(e[s]).modulo(t).getLowBitsUnsigned();
  return n;
}
var AC = Et((e, t) => e - t);
var DG = ov((e, t, n, s) => ({ real: e - n, imag: t - s }));
var fv = Ht(hi, AC, DG);
var FG = { kernelName: hi, backendName: "cpu", kernelFunc: fv };
function EC(e, t) {
  let n = new Array(e.rank);
  for (let r = 0; r < n.length; r++)
    n[r] = e.shape[r] * t[r];
  let s = Ae(n, e.dtype);
  for (let r = 0; r < s.values.length; ++r) {
    let a = s.indexToLoc(r), i = new Array(e.rank);
    for (let u = 0; u < i.length; u++)
      i[u] = a[u] % e.shape[u];
    let o = e.locToIndex(i);
    s.values[r] = e.values[o];
  }
  return s;
}
var Pu = (e, t) => {
  let n = t.value - e.value;
  return n === 0 ? e.index - t.index : n;
};
function RC(e, t, n = 0, s = e.length - 1) {
  for (; s > n; ) {
    if (s - n > 600) {
      let o = s - n + 1, u = t - n + 1, l = Math.log(o), c = 0.5 * Math.exp(2 * l / 3), p = 0.5 * Math.sqrt(l * c * (o - c) / o) * Math.sign(u - o / 2), d = Math.max(n, Math.floor(t - u * c / o + p)), h = Math.min(s, Math.floor(t + (o - u) * c / o + p));
      RC(e, t, d, h);
    }
    let r = e[t], a = n, i = s;
    for (w.swap(e, n, t), Pu(e[s], r) > 0 && w.swap(e, n, s); a < i; ) {
      for (w.swap(e, a, i), a++, i--; Pu(e[a], r) < 0; )
        a = a + 1;
      for (; Pu(e[i], r) > 0; )
        i = i - 1;
    }
    Pu(e[n], r) === 0 ? w.swap(e, n, i) : (i = i + 1, w.swap(e, i, s)), i <= t && (n = i + 1), t <= i && (s = i - 1);
  }
}
function DC(e, t, n, s, r) {
  let a = t[t.length - 1], [i, o] = [e.length / a, a], u = w.getTypedArrayFromDType(n, i * s), l = w.getTypedArrayFromDType("int32", i * s);
  for (let p = 0; p < i; p++) {
    let d = p * o, h = e.subarray(d, d + o), f = new Array(h.length);
    h.forEach((y, v) => f[v] = { value: y, index: v }), s < f.length && (RC(f, s), f = f.slice(0, s)), r && f.sort(Pu);
    let m = p * s, g = u.subarray(m, m + s), b = l.subarray(m, m + s);
    for (let y = 0; y < s; y++)
      g[y] = f[y].value, b[y] = f[y].index;
  }
  let c = t.slice();
  return c[c.length - 1] = s, [Ae(c, n, u), Ae(c, "int32", l)];
}
function FC(e, t, n, s) {
  let r = w.parseAxisParam(t, n)[0], a = [1, n[0], 1];
  for (let f = 0; f < r; f++)
    a[0] *= n[f];
  a[1] = n[r];
  for (let f = r + 1; f < n.length; f++)
    a[2] *= n[f];
  let i = {}, o = new Int32Array(n[r]), u = new Wt(a, s, e), l = [], c = a[0] === 1 && a[2] === 1;
  for (let f = 0; f < n[r]; f++) {
    let m;
    if (c)
      m = e[f].toString();
    else {
      let g = [];
      for (let b = 0; b < a[0]; b++)
        for (let y = 0; y < a[2]; y++)
          g.push(u.get(b, f, y));
      m = g.join(",");
    }
    if (i[m] !== void 0)
      o[f] = i[m];
    else {
      let g = Object.keys(i).length;
      i[m] = g, o[f] = g, l.push(f);
    }
  }
  let p = a.slice();
  p[1] = Object.keys(i).length;
  let d = new Wt(p, s);
  l.forEach((f, m) => {
    for (let g = 0; g < a[0]; g++)
      for (let b = 0; b < a[2]; b++)
        d.set(u.get(g, f, b), g, m, b);
  });
  let h = n.slice();
  return h[r] = p[1], { outputValues: d.values, outputShape: h, indices: o };
}
var hhe = "0.0.0";
vp("cpu", () => new Y0(), 1);
var OC = st(Pa, (e) => e >= 0 ? e : Math.exp(e) - 1);
var OG = { kernelName: Pa, backendName: "cpu", kernelFunc: OC };
function PC(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { alpha: a } = s;
  be([r], "leakyRelu");
  let i = w.sizeFromShape(r.shape), o = n.data.get(r.dataId).values, u = w.getTypedArrayFromDType("float32", i);
  for (let l = 0; l < o.length; l++)
    u[l] = o[l] < 0 ? a * o[l] : o[l];
  return n.makeTensorInfo(r.shape, "float32", u);
}
var PG = { kernelName: Ua, backendName: "cpu", kernelFunc: PC };
var zG = Et((e, t) => e < 0 ? t * e : e);
function zC(e) {
  let { inputs: t, backend: n } = e, { x: s, alpha: r } = t;
  be([s, r], "prelu");
  let a = n.data.get(s.dataId).values, i = n.data.get(r.dataId).values, [o, u] = zG(s.shape, r.shape, a, i, "float32");
  return n.makeTensorInfo(u, "float32", o);
}
var MG = { kernelName: ti, backendName: "cpu", kernelFunc: zC };
var MC = st(si, (e) => Math.max(0, e));
var LG = { kernelName: si, backendName: "cpu", kernelFunc: MC };
var LC = st(ai, (e) => Math.min(Math.max(0, e), 6));
var BG = { kernelName: ai, backendName: "cpu", kernelFunc: LC };
function Vd(e, t, n, s, r) {
  if (n === "linear")
    return Os({ inputs: { x: t }, backend: e });
  if (n === "relu")
    return MC({ inputs: { x: t }, backend: e });
  if (n === "elu")
    return OC({ inputs: { x: t }, backend: e });
  if (n === "relu6")
    return LC({ inputs: { x: t }, backend: e });
  if (n === "prelu")
    return zC({ inputs: { x: t, alpha: s }, backend: e });
  if (n === "leakyrelu")
    return PC({ inputs: { x: t }, backend: e, attrs: { alpha: r } });
  if (n === "sigmoid")
    return kC({ inputs: { x: t }, backend: e });
  throw new Error(`Activation ${n} has not been implemented for the CPU backend.`);
}
function pt(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { shape: a } = s, i = w.sizeFromShape(r.shape), o = w.inferFromImplicitShape(a, i), u = w.sizeFromShape(o);
  w.assert(i === u, () => `The new shape (${o}) has ${u} elements and the old shape (${r.shape}) has ${i} elements. The new shape and old shape must have the same number of elements.`), n.incRef(r.dataId);
  let l = n.data.get(r.dataId);
  if (l.complexTensorInfos != null) {
    let c = l.complexTensorInfos.real, p = l.complexTensorInfos.imag;
    c.shape = o, p.shape = o;
  }
  return { dataId: r.dataId, shape: o, dtype: r.dtype };
}
var VG = { kernelName: Oo, backendName: "cpu", kernelFunc: pt };
function BC(e) {
  let { inputs: t, backend: n, attrs: s } = e, { a: r, b: a } = t, { transposeA: i, transposeB: o } = s;
  be([r, a], "matMul");
  let u = r.shape.length, l = a.shape.length, c = i ? r.shape[u - 2] : r.shape[u - 1], p = o ? a.shape[l - 1] : a.shape[l - 2], d = i ? r.shape[u - 1] : r.shape[u - 2], h = o ? a.shape[l - 2] : a.shape[l - 1], f = r.shape.slice(0, -2), m = a.shape.slice(0, -2), g = w.sizeFromShape(f), b = w.sizeFromShape(m), v = Qo.assertAndGetBroadcastShape(r.shape.slice(0, -2), a.shape.slice(0, -2)).concat([d, h]);
  w.assert(c === p, () => `Error in matMul: inner shapes (${c}) and (${p}) of Tensors with shapes ${r.shape} and ${a.shape} and transposeA=${i} and transposeB=${o} must match.`);
  let x = i ? [g, c, d] : [g, d, c], k = o ? [b, h, p] : [b, p, h], I = pt({ inputs: { x: r }, backend: n, attrs: { shape: x } }), $ = pt({ inputs: { x: a }, backend: n, attrs: { shape: k } }), R = i ? I.shape[1] : I.shape[2], E = i ? I.shape[2] : I.shape[1], P = o ? $.shape[1] : $.shape[2], A = Math.max(g, b), O = n.data.get(I.dataId).values, T = n.data.get($.dataId).values, z = w.computeStrides(I.shape), W = w.computeStrides($.shape), [q, X, Y] = i ? [z[0], 1, z[1]] : [z[0], z[1], 1], [Z, te, J] = o ? [1, W[1], W[0]] : [W[1], 1, W[0]], se = E * P, ne = Ae([A, E, P], I.dtype), oe = ne.values, ae = n.blockSize;
  for (let de = 0; de < A; de++)
    for (let me = 0; me < E; me += ae)
      for (let ke = 0; ke < P; ke += ae)
        for (let Ie = 0; Ie < R; Ie += ae) {
          let Re = Math.min(me + ae, E), Pe = Math.min(ke + ae, P), Xe = Math.min(Ie + ae, R);
          for (let Je = me; Je < Re; Je++)
            for (let Ye = ke; Ye < Pe; Ye++) {
              let tt = 0;
              for (let Ce = Ie; Ce < Xe; Ce++) {
                let ut = Math.min(de, g - 1) * q, at = Math.min(de, b - 1) * J, Jt = O[ut + Je * X + Ce * Y], Nt = T[Ce * Z + Ye * te + at];
                tt += Jt * Nt;
              }
              oe[de * se + (Je * P + Ye)] += tt;
            }
        }
  return n.disposeIntermediateTensorInfo(I), n.disposeIntermediateTensorInfo($), n.makeTensorInfo(v, ne.dtype, ne.values);
}
var WG = { kernelName: Na, backendName: "cpu", kernelFunc: BC };
function UG(e) {
  let { inputs: t, backend: n, attrs: s } = e, { a: r, b: a, bias: i, preluActivationWeights: o } = t, { transposeA: u, transposeB: l, activation: c, leakyreluAlpha: p } = s, d, h, f, m = [];
  d = BC({ inputs: { a: r, b: a }, attrs: { transposeA: u, transposeB: l }, backend: n }), i && (h = io({ inputs: { a: d, b: i }, backend: n }), m.push(d), d = h), c && (f = Vd(n, d, c, o, p), m.push(d), d = f);
  for (let b of m)
    n.disposeIntermediateTensorInfo(b);
  return d;
}
var GG = { kernelName: aa, backendName: "cpu", kernelFunc: UG };
var HG = st(ol, (e) => Math.acos(e));
var qG = { kernelName: ol, backendName: "cpu", kernelFunc: HG };
var jG = st(ul, (e) => Math.acosh(e));
var KG = { kernelName: ul, backendName: "cpu", kernelFunc: jG };
function XG(e) {
  let { inputs: t, backend: n } = e, s = t;
  be(t, "addN");
  let r = s.map((o) => n.data.get(o.dataId).values), a = Ae(s[0].shape, s[0].dtype), i = a.values;
  for (let o = 0; o < s.length; o++) {
    let u = r[o];
    for (let l = 0; l < i.length; l++)
      i[l] += u[l];
  }
  return n.makeTensorInfo(a.shape, a.dtype, a.values);
}
var YG = { kernelName: Sa, backendName: "cpu", kernelFunc: XG };
function QG(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  be(r, "all");
  let o = w.parseAxisParam(a, r.shape), u = o, l = C.getAxesPermutation(u, r.shape.length), c = r;
  l != null && (c = wn({ inputs: { x: r }, backend: n, attrs: { perm: l } }), u = C.getInnerMostAxes(u.length, r.shape.length)), C.assertAxesAreInnerMostDims("all", u, c.shape.length);
  let [p, d] = C.computeOutAndReduceShapes(c.shape, u), h = w.sizeFromShape(d), f = w.makeZerosTypedArray(w.sizeFromShape(p), c.dtype), m = n.data.get(c.dataId).values;
  for (let b = 0; b < f.length; ++b) {
    let y = b * h, v = m[y];
    for (let x = 0; x < h; ++x) {
      let k = m[y + x];
      v = v && k;
    }
    f[b] = v;
  }
  l != null && n.disposeIntermediateTensorInfo(c);
  let g = n.makeTensorInfo(p, c.dtype, f);
  if (i) {
    let b = C.expandShapeToKeepDim(p, o), y = pt({ inputs: { x: g }, backend: n, attrs: { shape: b } });
    return n.disposeIntermediateTensorInfo(g), y;
  }
  return g;
}
var ZG = { kernelName: ll, backendName: "cpu", kernelFunc: QG };
function JG(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  be(r, "any");
  let o = w.parseAxisParam(a, r.shape), u = o, l = C.getAxesPermutation(u, r.shape.length), c = r;
  l != null && (c = wn({ inputs: { x: r }, backend: n, attrs: { perm: l } }), u = C.getInnerMostAxes(u.length, r.shape.length)), C.assertAxesAreInnerMostDims("any", u, c.shape.length);
  let [p, d] = C.computeOutAndReduceShapes(c.shape, u), h = w.sizeFromShape(d), f = w.makeZerosTypedArray(w.sizeFromShape(p), c.dtype), m = n.data.get(c.dataId).values;
  for (let b = 0; b < f.length; ++b) {
    let y = b * h, v = m[y];
    for (let x = 0; x < h; ++x) {
      let k = m[y + x];
      v = v || k;
    }
    f[b] = v;
  }
  l != null && n.disposeIntermediateTensorInfo(c);
  let g = n.makeTensorInfo(p, c.dtype, f);
  if (i) {
    let b = C.expandShapeToKeepDim(p, o), y = pt({ inputs: { x: g }, backend: n, attrs: { shape: b } });
    return n.disposeIntermediateTensorInfo(g), y;
  }
  return g;
}
var eH = { kernelName: cl, backendName: "cpu", kernelFunc: JG };
function tH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a } = s;
  be(r, "argMax");
  let i = w.parseAxisParam(a, r.shape), o = C.getAxesPermutation(i, r.shape.length), u = r, l = [];
  o != null && (u = wn({ inputs: { x: r }, backend: n, attrs: { perm: o } }), l.push(u), i = C.getInnerMostAxes(i.length, u.shape.length)), i = [i[0]], C.assertAxesAreInnerMostDims("argMax", i, u.shape.length);
  let [c, p] = C.computeOutAndReduceShapes(u.shape, i), d = w.sizeFromShape(c), h = w.makeZerosTypedArray(d, "int32"), f = w.sizeFromShape(p), m = n.data.get(u.dataId).values;
  for (let g = 0; g < h.length; ++g) {
    let b = g * f, y = m[b], v = 0;
    for (let x = 0; x < f; ++x) {
      let k = m[b + x];
      k > y && (y = k, v = x);
    }
    h[g] = v;
  }
  return l.forEach((g) => n.disposeIntermediateTensorInfo(g)), n.makeTensorInfo(c, "int32", h);
}
var nH = { kernelName: Ia, backendName: "cpu", kernelFunc: tH };
function sH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a } = s;
  be(r, "argMin");
  let i = w.parseAxisParam(a, r.shape), o = C.getAxesPermutation(i, r.shape.length), u = r, l = [];
  o != null && (u = wn({ inputs: { x: r }, backend: n, attrs: { perm: o } }), l.push(u), i = C.getInnerMostAxes(i.length, u.shape.length)), i = [i[0]], C.assertAxesAreInnerMostDims("argMin", i, u.shape.length);
  let [c, p] = C.computeOutAndReduceShapes(u.shape, i), d = w.sizeFromShape(c), h = w.makeZerosTypedArray(d, "int32"), f = w.sizeFromShape(p), m = n.data.get(u.dataId).values;
  for (let g = 0; g < h.length; ++g) {
    let b = g * f, y = m[b], v = 0;
    for (let x = 0; x < f; ++x) {
      let k = m[b + x];
      k < y && (y = k, v = x);
    }
    h[g] = v;
  }
  return l.forEach((g) => n.disposeIntermediateTensorInfo(g)), n.makeTensorInfo(c, "int32", h);
}
var rH = { kernelName: dl, backendName: "cpu", kernelFunc: sH };
var aH = st(pl, (e) => Math.asin(e));
var iH = { kernelName: pl, backendName: "cpu", kernelFunc: aH };
var oH = st(hl, (e) => Math.asinh(e));
var uH = { kernelName: hl, backendName: "cpu", kernelFunc: oH };
var lH = st(fl, (e) => Math.atan(e));
var cH = { kernelName: fl, backendName: "cpu", kernelFunc: lH };
var dH = Et((e, t) => Math.atan2(e, t));
var pH = Ht(gl, dH);
var hH = { kernelName: gl, backendName: "cpu", kernelFunc: pH };
var fH = st(ml, (e) => Math.atanh(e));
var mH = { kernelName: ml, backendName: "cpu", kernelFunc: fH };
function mv(e, t, n, s, r, a) {
  let i = r.strideHeight, o = r.strideWidth, u = r.dilationHeight, l = r.dilationWidth, c = r.effectiveFilterHeight, p = r.effectiveFilterWidth, d = r.padInfo.top, h = r.padInfo.left, f = a === "max" ? Number.NEGATIVE_INFINITY : Number.POSITIVE_INFINITY, m = Ae(r.outShape, n), g = m.values, b = r.outShape[1] * r.outShape[2] * r.outShape[3], y = r.outShape[2] * r.outShape[3], v = r.outShape[3];
  for (let x = 0; x < r.batchSize; ++x) {
    let k = x * b, I = x * s[0];
    for (let $ = 0; $ < r.inChannels; ++$)
      for (let R = 0; R < r.outHeight; ++R) {
        let E = R * i - d, P = Math.max(0, E), A = Math.min(r.inHeight, c + E), O = k + R * y;
        for (let T = 0; T < r.outWidth; ++T) {
          let z = T * o - h, W = Math.max(0, z), q = Math.min(r.inWidth, p + z), X = f, Y = 0, Z = 0;
          for (let J = P; J < A; J += u) {
            let se = I + J * s[1];
            for (let ne = W; ne < q; ne += l) {
              let oe = se + ne * s[2], ae = e[oe + $];
              a === "max" && ae > X ? X = ae : a === "avg" && (Y += ae, Z++);
            }
            if (isNaN(X))
              break;
          }
          let te = O + T * v + $;
          g[te] = a === "avg" ? Y / Z : X;
        }
      }
  }
  return m;
}
function VC(e, t, n, s, r = false, a = false) {
  let i = Ae(s.outShape, "int32"), o = s.strideHeight, u = s.strideWidth, l = s.dilationHeight, c = s.dilationWidth, p = s.effectiveFilterHeight, d = s.effectiveFilterWidth, h = s.padInfo.top, f = s.padInfo.left, m = Ae(t, n, e);
  for (let g = 0; g < s.batchSize; ++g)
    for (let b = 0; b < s.inChannels; ++b)
      for (let y = 0; y < s.outHeight; ++y) {
        let v = y * o - h, x = v;
        for (; x < 0; )
          x += l;
        let k = Math.min(s.inHeight, p + v);
        for (let I = 0; I < s.outWidth; ++I) {
          let $ = I * u - f, R = $;
          for (; R < 0; )
            R += c;
          let E = Math.min(s.inWidth, d + $), P = Number.NEGATIVE_INFINITY, A = -1;
          for (let O = x; O < k; O += l) {
            let T = O - v;
            for (let z = R; z < E; z += c) {
              let W = z - $, q = m.get(g, O, z, b);
              q > P && (P = q, r ? A = a ? ((g * s.inHeight + O) * s.inWidth + z) * s.inChannels + b : (O * s.inWidth + z) * s.inChannels + b : A = T * d + W);
            }
          }
          i.set(A, g, y, I, b);
        }
      }
  return i;
}
function WC(e, t, n, s, r, a) {
  let i = r.strideDepth, o = r.strideHeight, u = r.strideWidth, l = r.dilationDepth, c = r.dilationHeight, p = r.dilationWidth, d = r.effectiveFilterDepth, h = r.effectiveFilterHeight, f = r.effectiveFilterWidth, m = r.padInfo.front, g = r.padInfo.top, b = r.padInfo.left, y = a === "max" ? Number.NEGATIVE_INFINITY : Number.POSITIVE_INFINITY, v = Ae(r.outShape, n), x = v.values, k = r.outShape[1] * r.outShape[2] * r.outShape[3] * r.outShape[4], I = r.outShape[2] * r.outShape[3] * r.outShape[4], $ = r.outShape[3] * r.outShape[4], R = r.outShape[4];
  for (let E = 0; E < r.batchSize; ++E) {
    let P = E * k, A = E * s[0];
    for (let O = 0; O < r.inChannels; ++O)
      for (let T = 0; T < r.outDepth; ++T) {
        let z = T * i - m, W = z;
        for (; W < 0; )
          W += l;
        let q = Math.min(r.inDepth, d + z), X = P + T * I;
        for (let Y = 0; Y < r.outHeight; ++Y) {
          let Z = Y * o - g, te = Z;
          for (; te < 0; )
            te += c;
          let J = Math.min(r.inHeight, h + Z), se = X + Y * $;
          for (let ne = 0; ne < r.outWidth; ++ne) {
            let oe = ne * u - b, ae = oe;
            for (; ae < 0; )
              ae += p;
            let de = Math.min(r.inWidth, f + oe), me = se + ne * R, ke = y, Ie = 0, Re = 0;
            for (let Xe = W; Xe < q; Xe += l) {
              let Je = A + Xe * s[1];
              for (let Ye = te; Ye < J; Ye += c) {
                let tt = Je + Ye * s[2];
                for (let Ce = ae; Ce < de; Ce += p) {
                  let ut = tt + Ce * s[3], at = e[ut + O];
                  if (a === "max" && at > ke ? ke = at : a === "avg" && (Ie += at, Re++), isNaN(ke))
                    break;
                }
                if (isNaN(ke))
                  break;
              }
              if (isNaN(ke))
                break;
            }
            let Pe = me + O;
            x[Pe] = a === "avg" ? Ie / Re : ke;
          }
        }
      }
  }
  return v;
}
function gH(e, t) {
  let n = Ae(t.outShape, "int32"), s = t.strideDepth, r = t.strideHeight, a = t.strideWidth, i = t.dilationDepth, o = t.dilationHeight, u = t.dilationWidth, l = t.effectiveFilterDepth, c = t.effectiveFilterHeight, p = t.effectiveFilterWidth, d = t.padInfo.front, h = t.padInfo.top, f = t.padInfo.left;
  for (let m = 0; m < t.batchSize; ++m)
    for (let g = 0; g < t.inChannels; ++g)
      for (let b = 0; b < t.outDepth; ++b) {
        let y = b * s - d, v = y;
        for (; v < 0; )
          v += i;
        let x = Math.min(t.inDepth, l + y);
        for (let k = 0; k < t.outHeight; ++k) {
          let I = k * r - h, $ = I;
          for (; $ < 0; )
            $ += o;
          let R = Math.min(t.inHeight, c + I);
          for (let E = 0; E < t.outWidth; ++E) {
            let P = E * a - f, A = P;
            for (; A < 0; )
              A += u;
            let O = Math.min(t.inWidth, p + P), T = Number.NEGATIVE_INFINITY, z = -1;
            for (let W = v; W < x; W += i) {
              let q = W - y;
              for (let X = $; X < R; X += o) {
                let Y = X - I;
                for (let Z = A; Z < O; Z += u) {
                  let te = Z - P, J = e.get(m, W, X, Z, g);
                  J >= T && (T = J, z = q * c * p + Y * c + te);
                }
              }
            }
            n.set(z, m, b, k, E, g);
          }
        }
      }
  return n;
}
function bH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t;
  be(r, "avgPool");
  let { filterSize: a, strides: i, pad: o, dimRoundingMode: u } = s, l = 1;
  w.assert(C.eitherStridesOrDilationsAreOne(i, l), () => `Error in avgPool: Either strides or dilations must be 1. Got strides ${i} and dilations '${l}'`);
  let c = C.computePool2DInfo(r.shape, a, i, l, o, u), p;
  if (c.filterWidth === 1 && c.filterHeight === 1 && w.arraysEqual(c.inShape, c.outShape))
    p = Os({ inputs: { x: r }, backend: n });
  else {
    let d = n.data.get(r.dataId).values, h = w.computeStrides(r.shape), f = mv(d, r.shape, r.dtype, h, c, "avg");
    p = n.makeTensorInfo(c.outShape, r.dtype, f.values);
  }
  return p;
}
var yH = { kernelName: Ca, backendName: "cpu", kernelFunc: bH };
function vH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { filterSize: a, strides: i, pad: o, dimRoundingMode: u, dataFormat: l } = s;
  be(r, "avgPool3d");
  let c = C.computePool3DInfo(r.shape, a, i, 1, o, u, l), p = n.data.get(r.dataId).values, d = WC(p, r.shape, r.dtype, w.computeStrides(r.shape), c, "avg");
  return n.makeTensorInfo(d.shape, "float32", d.values);
}
var xH = { kernelName: Jd, backendName: "cpu", kernelFunc: vH };
function wH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, input: a } = t, { filterSize: i, strides: o, pad: u, dimRoundingMode: l } = s;
  be([r, a], "avgPool3DGrad");
  let c = C.computePool3DInfo(a.shape, i, o, 1, u, l), p = c.strideDepth, d = c.strideHeight, h = c.strideWidth, f = c.filterDepth, m = c.filterHeight, g = c.filterWidth, b = c.dilationDepth, y = c.dilationHeight, v = c.dilationWidth, x = c.effectiveFilterDepth, k = c.effectiveFilterHeight, I = c.effectiveFilterWidth, $ = x - 1 - c.padInfo.front, R = I - 1 - c.padInfo.left, E = k - 1 - c.padInfo.top, P = Ae(a.shape, "float32"), A = 1 / (f * m * g), O = n.bufferSync(r);
  for (let T = 0; T < c.batchSize; ++T)
    for (let z = 0; z < c.inChannels; ++z)
      for (let W = 0; W < c.inDepth; ++W)
        for (let q = 0; q < c.inHeight; ++q)
          for (let X = 0; X < c.inWidth; ++X) {
            let Y = W - $, Z = q - E, te = X - R, J = 0;
            for (let se = 0; se < x; se += b) {
              let ne = (Y + se) / p;
              if (!(ne < 0 || ne >= c.outDepth || Math.floor(ne) !== ne))
                for (let oe = 0; oe < k; oe += y) {
                  let ae = (Z + oe) / d;
                  if (!(ae < 0 || ae >= c.outHeight || Math.floor(ae) !== ae))
                    for (let de = 0; de < I; de += v) {
                      let me = (te + de) / h;
                      if (me < 0 || me >= c.outWidth || Math.floor(me) !== me)
                        continue;
                      J += O.get(T, ne, ae, me, z);
                    }
                }
            }
            P.set(J * A, T, W, q, X, z);
          }
  return n.makeTensorInfo(P.shape, P.dtype, P.values);
}
var kH = { kernelName: hg, backendName: "cpu", kernelFunc: wH };
function SH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, input: a } = t, i = a;
  be([r, a], "avgPoolGrad");
  let { filterSize: o, strides: u, pad: l } = s, c = C.computePool2DInfo(i.shape, o, u, 1, l), p = c.strideHeight, d = c.strideWidth, h = c.filterHeight, f = c.filterWidth, m = c.dilationHeight, g = c.dilationWidth, b = c.effectiveFilterHeight, y = c.effectiveFilterWidth, v = y - 1 - c.padInfo.left, x = b - 1 - c.padInfo.top, k = Ae(i.shape, "float32"), I = 1 / (h * f), $ = n.data.get(r.dataId).values, R = Ae(r.shape, "float32", $);
  for (let E = 0; E < c.batchSize; ++E)
    for (let P = 0; P < c.inChannels; ++P)
      for (let A = 0; A < c.inHeight; ++A)
        for (let O = 0; O < c.inWidth; ++O) {
          let T = A - x, z = O - v, W = 0;
          for (let q = 0; q < b; q += m) {
            let X = (T + q) / p;
            if (!(X < 0 || X >= c.outHeight || Math.floor(X) !== X))
              for (let Y = 0; Y < y; Y += g) {
                let Z = (z + Y) / d;
                if (Z < 0 || Z >= c.outWidth || Math.floor(Z) !== Z)
                  continue;
                W += R.get(E, X, Z, P);
              }
          }
          k.set(W * I, E, A, O, P);
        }
  return n.makeTensorInfo(k.shape, k.dtype, k.values);
}
var IH = { kernelName: pg, backendName: "cpu", kernelFunc: SH };
function CH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, scale: a, offset: i, mean: o, variance: u } = t;
  w.assert(o.shape.length === u.shape.length, () => "Batch normalization gradient requires mean and variance to have equal ranks."), w.assert(i == null || o.shape.length === i.shape.length, () => "Batch normalization gradient requires mean and offset to have equal ranks."), w.assert(a == null || o.shape.length === a.shape.length, () => "Batch normalization gradient requires mean and scale to have equal ranks."), be([r, o, u, a, i], "batchNorm");
  let { varianceEpsilon: l } = s;
  l == null && (l = 1e-3);
  let c = n.data.get(r.dataId).values, p = n.data.get(o.dataId).values, d = n.data.get(u.dataId).values, h = a ? n.data.get(a.dataId).values : new Float32Array([1]), f = i ? n.data.get(i.dataId).values : new Float32Array([0]), m = new Float32Array(c.length), g = f.length, b = h.length, y = d.length, v = p.length, x = 0, k = 0, I = 0, $ = 0;
  for (let R = 0; R < c.length; ++R)
    m[R] = f[x++] + (c[R] - p[k++]) * h[I++] / Math.sqrt(d[$++] + l), x >= g && (x = 0), k >= v && (k = 0), I >= b && (I = 0), $ >= y && ($ = 0);
  return n.makeTensorInfo(r.shape, r.dtype, m);
}
var NH = { kernelName: Ba, backendName: "cpu", kernelFunc: CH };
function TH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockShape: a, crops: i } = s;
  be([r], "batchToSpaceND");
  let o = a.reduce((b, y) => b * y), u = C.getReshaped(r.shape, a, o), l = C.getPermuted(u.length, a.length), c = C.getReshapedPermuted(r.shape, a, o), p = C.getSliceBeginCoords(i, a.length), d = C.getSliceSize(c, i, a.length), h = pt({ inputs: { x: r }, backend: n, attrs: { shape: u } }), f = wn({ inputs: { x: h }, backend: n, attrs: { perm: l } }), m = pt({ inputs: { x: f }, backend: n, attrs: { shape: c } }), g = ba({ inputs: { x: m }, backend: n, attrs: { begin: p, size: d } });
  return n.disposeIntermediateTensorInfo(h), n.disposeIntermediateTensorInfo(f), n.disposeIntermediateTensorInfo(m), g;
}
var $H = { kernelName: ho, backendName: "cpu", kernelFunc: TH };
function _H(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, weights: a } = t, { size: i } = s, o = n.data.get(r.dataId).values, u = n.data.get(a.dataId).values, l = uv(o, u, a.dtype, a.shape, i);
  return n.makeTensorInfo([i], a.dtype, l);
}
var AH = { kernelName: fg, backendName: "cpu", kernelFunc: _H };
function EH(e) {
  let { inputs: t, backend: n } = e, { s0: s, s1: r } = t, a = n.data.get(s.dataId).values, i = n.data.get(r.dataId).values, o = C.assertAndGetBroadcastShape(Array.from(a), Array.from(i));
  return n.makeTensorInfo([o.length], "int32", Int32Array.from(o));
}
var RH = { kernelName: mg, backendName: "cpu", kernelFunc: EH };
var DH = st(Ir, (e, t) => {
  let n = t;
  return e > n.clipValueMax ? n.clipValueMax : e < n.clipValueMin ? n.clipValueMin : e;
});
var FH = { kernelName: Ir, backendName: "cpu", kernelFunc: DH };
var OH = (e) => {
  let { x: t } = e.inputs, n = e.backend, s = new Float32Array(w.sizeFromShape(t.shape)), r = n.data.get(t.dataId), a = r.complexTensorInfos.real, i = r.complexTensorInfos.imag, o = n.data.get(a.dataId).values, u = n.data.get(i.dataId).values;
  for (let l = 0; l < o.length; l++) {
    let c = o[l], p = u[l];
    s[l] = Math.hypot(c, p);
  }
  return n.makeOutput(s, t.shape, "float32");
};
var PH = { kernelName: tp, backendName: "cpu", kernelFunc: OH };
function oo(e) {
  let { inputs: t, backend: n } = e, { input: s } = t, r = n.data.get(s.dataId).complexTensorInfos.imag, a = n.data.get(r.dataId).values;
  return n.makeTensorInfo(r.shape, r.dtype, a);
}
var zH = { kernelName: ap, backendName: "cpu", kernelFunc: oo };
function uo(e) {
  let { inputs: t, backend: n, attrs: s } = e, { axis: r } = s, a = w.parseAxisParam(r, t[0].shape)[0], i = C.computeOutShape(t.map((m) => m.shape), a);
  if (w.sizeFromShape(i) === 0)
    return n.makeTensorInfo(i, t[0].dtype, []);
  let o = t.filter((m) => w.sizeFromShape(m.shape) > 0);
  if (o.length === 1)
    return Os({ inputs: { x: o[0] }, backend: n });
  let u = o.map((m) => m.shape);
  if (C.assertParamsConsistent(u, a), o[0].dtype === "complex64") {
    let m = o.map((x) => ga({ inputs: { input: x }, backend: n })), g = o.map((x) => oo({ inputs: { input: x }, backend: n })), b = uo({ inputs: m, backend: n, attrs: { axis: a } }), y = uo({ inputs: g, backend: n, attrs: { axis: a } }), v = En({ inputs: { real: b, imag: y }, backend: n });
    return m.forEach((x) => n.disposeIntermediateTensorInfo(x)), g.forEach((x) => n.disposeIntermediateTensorInfo(x)), n.disposeIntermediateTensorInfo(b), n.disposeIntermediateTensorInfo(y), v;
  }
  let l = o.map((m) => {
    let g = w.sizeFromShape(m.shape.slice(a));
    return pt({ inputs: { x: m }, backend: n, attrs: { shape: [-1, g] } });
  }), c = l.map((m) => ({ vals: n.data.get(m.dataId).values, shape: m.shape }));
  i = C.computeOutShape(l.map((m) => m.shape), 1);
  let p = l[0].shape[0] === 1, d = lv(c, i, t[0].dtype, p), h = C.computeOutShape(o.map((m) => m.shape), a), f = n.makeTensorInfo(h, t[0].dtype, d);
  return l.forEach((m) => n.disposeIntermediateTensorInfo(m)), f;
}
var MH = { kernelName: fo, backendName: "cpu", kernelFunc: uo };
function UC(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dataFormat: u, dilations: l, dimRoundingMode: c } = s;
  be([r, a], "conv2d");
  let p = C.convertConv2DDataFormat(u), d = C.computeConv2DInfo(r.shape, a.shape, i, l, o, c, false, p), h = d.filterHeight, f = d.filterWidth, m = d.dilationHeight, g = d.dilationWidth, b = d.padInfo.left, y = d.padInfo.top, v = d.dataFormat === "channelsLast", x = new Wt(d.outShape, r.dtype), k = w.computeStrides(r.shape), I = w.computeStrides(a.shape), $ = k[0], R = v ? k[1] : k[2], E = v ? k[2] : 1, P = v ? 1 : k[1], A = x.strides[0], O = v ? x.strides[1] : x.strides[2], T = v ? x.strides[2] : 1, z = v ? 1 : x.strides[1], W = n.data.get(r.dataId).values, q = n.data.get(a.dataId).values, X = x.values;
  for (let Y = 0; Y < d.batchSize; ++Y) {
    let Z = Y * $, te = Y * A;
    for (let J = 0; J < d.outHeight; ++J) {
      let se = te + J * O, ne = J * d.strideHeight - y;
      for (let oe = 0; oe < h; ++oe) {
        let ae = ne + oe * m;
        if (ae < 0 || ae >= d.inHeight)
          continue;
        let de = oe * I[0], me = Z + ae * R;
        for (let ke = 0; ke < d.outWidth; ++ke) {
          let Ie = se + ke * T, Re = ke * d.strideWidth - b;
          for (let Pe = 0; Pe < f; ++Pe) {
            let Xe = Re + Pe * g;
            if (Xe < 0 || Xe >= d.inWidth)
              continue;
            let Je = de + Pe * I[1], Ye = me + Xe * E, tt = Je;
            for (let Ce = 0; Ce < d.inChannels; ++Ce) {
              let ut = W[Ye + Ce * P];
              for (let at = 0; at < d.outChannels; ++at)
                X[Ie + at * z] += ut * q[tt + at];
              tt += d.outChannels;
            }
          }
        }
      }
    }
  }
  return n.makeTensorInfo(x.shape, x.dtype, X);
}
var LH = { kernelName: _a, backendName: "cpu", kernelFunc: UC };
function BH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, dy: a } = t, { strides: i, pad: o, dataFormat: u, dimRoundingMode: l, filterShape: c } = s;
  be([r, a], "conv2dBackpropFilter");
  let p = C.convertConv2DDataFormat(u), d = C.computeConv2DInfo(r.shape, c, i, 1, o, l, false, p), { strideHeight: h, strideWidth: f, filterHeight: m, filterWidth: g } = d, b = d.dataFormat === "channelsLast", y = new Wt(d.filterShape, "float32"), v = d.padInfo.left, x = d.padInfo.top, k = n.data.get(r.dataId).values, I = n.data.get(a.dataId).values, $ = new Wt(r.shape, r.dtype, k), R = new Wt(a.shape, a.dtype, I);
  for (let E = 0; E < m; ++E) {
    let P = Math.max(0, Math.ceil((x - E) / h)), A = Math.min(d.outHeight, (d.inHeight + x - E) / h);
    for (let O = 0; O < g; ++O) {
      let T = Math.max(0, Math.ceil((v - O) / f)), z = Math.min(d.outWidth, (d.inWidth + v - O) / f);
      for (let W = 0; W < d.inChannels; ++W)
        for (let q = 0; q < d.outChannels; ++q) {
          let X = 0;
          for (let Y = 0; Y < d.batchSize; ++Y)
            for (let Z = P; Z < A; ++Z) {
              let te = E + Z * h - x;
              for (let J = T; J < z; ++J) {
                let se = O + J * f - v;
                b ? X += $.get(Y, te, se, W) * R.get(Y, Z, J, q) : X += $.get(Y, W, te, se) * R.get(Y, q, Z, J);
              }
            }
          y.set(X, E, O, W, q);
        }
    }
  }
  return n.makeTensorInfo(y.shape, y.dtype, y.values);
}
var VH = { kernelName: gg, backendName: "cpu", kernelFunc: BH };
function WH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, filter: a } = t, { inputShape: i, strides: o, pad: u, dataFormat: l, dimRoundingMode: c } = s;
  be([r, a], "conv2dBackpropInput");
  let p = w.computeStrides(a.shape), d = w.computeStrides(r.shape), h = C.convertConv2DDataFormat(l), f = C.computeConv2DInfo(i, a.shape, o, 1, u, c, false, h), m = new Wt(f.inShape, "float32"), g = m.values, b = n.data.get(r.dataId).values, y = n.data.get(a.dataId).values, [v, x, k] = p, { batchSize: I, filterHeight: $, filterWidth: R, inChannels: E, inHeight: P, inWidth: A, outChannels: O, outHeight: T, outWidth: z, strideHeight: W, strideWidth: q } = f;
  h = f.dataFormat;
  let X = $ - 1 - f.padInfo.top, Y = R - 1 - f.padInfo.left, Z = h === "channelsLast", te = m.strides[0], J = Z ? m.strides[1] : m.strides[2], se = Z ? m.strides[2] : 1, ne = Z ? 1 : m.strides[1], oe = d[0], ae = Z ? d[1] : d[2], de = Z ? d[2] : 1, me = Z ? 1 : d[1];
  for (let ke = 0; ke < I; ++ke)
    for (let Ie = 0; Ie < E; ++Ie)
      for (let Re = 0; Re < P; ++Re) {
        let Pe = Re - X, Xe = Math.max(0, Math.ceil(Pe / W)), Je = Math.min(T, ($ + Pe) / W);
        for (let Ye = 0; Ye < A; ++Ye) {
          let tt = Ye - Y, Ce = Math.max(0, Math.ceil(tt / q)), ut = Math.min(z, (R + tt) / q), at = 0;
          for (let Nt = Xe; Nt < Je; ++Nt) {
            let In = Nt * W - Pe;
            for (let Rt = Ce; Rt < ut; ++Rt) {
              let en = Rt * q - tt, Cn = oe * ke + ae * Nt + de * Rt, Nn = v * ($ - 1 - In) + x * (R - 1 - en) + k * Ie;
              for (let Yt = 0; Yt < O; ++Yt) {
                let Dn = b[Cn + me * Yt], tn = y[Nn + Yt];
                at += Dn * tn;
              }
            }
          }
          let Jt = te * ke + J * Re + se * Ye + ne * Ie;
          g[Jt] = at;
        }
      }
  return n.makeTensorInfo(m.shape, m.dtype, m.values);
}
var UH = { kernelName: Aa, backendName: "cpu", kernelFunc: WH };
function GH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dilations: u } = s;
  be([r, a], "conv3d");
  let l = C.computeConv3DInfo(r.shape, a.shape, i, u, o), { filterDepth: c, filterHeight: p, filterWidth: d, dilationDepth: h, dilationHeight: f, dilationWidth: m, padInfo: g } = l, b = g.front, y = g.left, v = g.top, x = new Wt(l.outShape, r.dtype), k = n.data.get(r.dataId).values, I = n.data.get(a.dataId).values, $ = x.values, R = w.computeStrides(r.shape), E = w.computeStrides(a.shape);
  for (let P = 0; P < l.batchSize; ++P) {
    let A = P * R[0], O = P * x.strides[0];
    for (let T = 0; T < l.outDepth; ++T) {
      let z = O + T * x.strides[1], W = T * l.strideDepth - b;
      for (let q = 0; q < c; ++q) {
        let X = W + q * h;
        if (X < 0 || X >= l.inDepth)
          continue;
        let Y = q * E[0], Z = A + X * R[1];
        for (let te = 0; te < l.outHeight; ++te) {
          let J = z + te * x.strides[2], se = te * l.strideHeight - v;
          for (let ne = 0; ne < p; ++ne) {
            let oe = se + ne * f;
            if (oe < 0 || oe >= l.inHeight)
              continue;
            let ae = Y + ne * E[1], de = Z + oe * R[2];
            for (let me = 0; me < l.outWidth; ++me) {
              let ke = J + me * l.outChannels, Ie = me * l.strideWidth - y;
              for (let Re = 0; Re < d; ++Re) {
                let Pe = Ie + Re * m;
                if (Pe < 0 || Pe >= l.inWidth)
                  continue;
                let Xe = ae + Re * E[2], Je = de + Pe * l.inChannels, Ye = Xe;
                for (let tt = 0; tt < l.inChannels; ++tt) {
                  let Ce = k[Je + tt];
                  for (let ut = 0; ut < l.outChannels; ++ut)
                    $[ke + ut] += Ce * I[Ye + ut];
                  Ye += l.outChannels;
                }
              }
            }
          }
        }
      }
    }
  }
  return n.makeTensorInfo(x.shape, x.dtype, x.values);
}
var HH = { kernelName: np, backendName: "cpu", kernelFunc: GH };
function qH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, dy: a } = t, { strides: i, pad: o, filterShape: u } = s;
  be([r, a], "conv3dBackpropFilterV2");
  let l = w.computeStrides(r.shape), c = w.computeStrides(a.shape), p = C.computeConv3DInfo(r.shape, u, i, 1, o), d = p.strideDepth, h = p.strideHeight, f = p.strideWidth, m = p.filterDepth, g = p.filterHeight, b = p.filterWidth, y = new Wt(p.filterShape, "float32"), v = y.values, [x, k, I, $] = y.strides, R = n.data.get(a.dataId).values, [E, P, A, O] = c, T = n.data.get(r.dataId).values, [z, W, q, X] = l, Y = p.padInfo.front, Z = p.padInfo.left, te = p.padInfo.top;
  for (let J = 0; J < m; ++J) {
    let se = Math.max(0, Math.ceil((Y - J) / d)), ne = Math.min(p.outDepth, (p.inDepth + Y - J) / d), oe = J * x;
    for (let ae = 0; ae < g; ++ae) {
      let de = Math.max(0, Math.ceil((te - ae) / h)), me = Math.min(p.outHeight, (p.inHeight + te - ae) / h), ke = ae * k + oe;
      for (let Ie = 0; Ie < b; ++Ie) {
        let Re = Math.max(0, Math.ceil((Z - Ie) / f)), Pe = Math.min(p.outWidth, (p.inWidth + Z - Ie) / f), Xe = Ie * I + ke;
        for (let Je = 0; Je < p.inChannels; ++Je) {
          let Ye = Je * $ + Xe;
          for (let tt = 0; tt < p.outChannels; ++tt) {
            let Ce = 0;
            for (let ut = 0; ut < p.batchSize; ++ut) {
              let at = ut * z, Jt = ut * E;
              for (let Nt = se; Nt < ne; ++Nt) {
                let Rt = (J + Nt * d - Y) * W + at, en = Nt * P + Jt;
                for (let Cn = de; Cn < me; ++Cn) {
                  let Yt = (ae + Cn * h - te) * q + Rt, Dn = Cn * A + en;
                  for (let tn = Re; tn < Pe; ++tn) {
                    let Ms = (Ie + tn * f - Z) * X + Yt, Ci = tn * O + Dn;
                    Ce += T[Ms + Je] * R[Ci + tt];
                  }
                }
              }
            }
            v[Ye + tt] = Ce;
          }
        }
      }
    }
  }
  return n.makeTensorInfo(y.shape, y.dtype, y.values);
}
var jH = { kernelName: bg, backendName: "cpu", kernelFunc: qH };
function KH(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, filter: a } = t, { pad: i, strides: o, inputShape: u } = s;
  be([r], "conv3dBackpropInputV2");
  let l = w.computeStrides(r.shape), c = w.computeStrides(a.shape), p = C.computeConv3DInfo(u, a.shape, o, 1, i), d = new Wt(p.inShape, "float32"), h = d.values, [f, m, g, b] = d.strides, y = n.data.get(r.dataId).values, [v, x, k, I] = l, $ = n.data.get(a.dataId).values, [R, E, P, A] = c, { batchSize: O, filterDepth: T, filterHeight: z, filterWidth: W, inChannels: q, inDepth: X, inHeight: Y, inWidth: Z, outChannels: te, outDepth: J, outHeight: se, outWidth: ne, strideDepth: oe, strideHeight: ae, strideWidth: de } = p, me = T - 1 - p.padInfo.front, ke = z - 1 - p.padInfo.top, Ie = W - 1 - p.padInfo.left;
  for (let Re = 0; Re < O; ++Re)
    for (let Pe = 0; Pe < q; ++Pe)
      for (let Xe = 0; Xe < X; ++Xe) {
        let Je = Xe - me, Ye = Math.max(0, Math.ceil(Je / oe)), tt = Math.min(J, (T + Je) / oe);
        for (let Ce = 0; Ce < Y; ++Ce) {
          let ut = Ce - ke, at = Math.max(0, Math.ceil(ut / ae)), Jt = Math.min(se, (z + ut) / ae);
          for (let Nt = 0; Nt < Z; ++Nt) {
            let In = Nt - Ie, Rt = Math.max(0, Math.ceil(In / de)), en = Math.min(ne, (W + In) / de), Cn = 0;
            for (let Nn = Ye; Nn < tt; ++Nn) {
              let Yt = Nn * oe - Je;
              for (let Dn = at; Dn < Jt; ++Dn) {
                let tn = Dn * ae - ut;
                for (let zs = Rt; zs < en; ++zs) {
                  let Ms = zs * de - In, Ci = v * Re + x * Nn + k * Dn + I * zs, Zs = R * (T - 1 - Yt) + E * (z - 1 - tn) + P * (W - 1 - Ms) + A * Pe;
                  for (let Ls = 0; Ls < te; ++Ls) {
                    let gu = y[Ci + Ls], Ni = $[Zs + Ls];
                    Cn += gu * Ni;
                  }
                }
              }
            }
            h[f * Re + m * Xe + g * Ce + b * Nt + Pe] = Cn;
          }
        }
      }
  return n.makeTensorInfo(d.shape, d.dtype, d.values);
}
var XH = { kernelName: yg, backendName: "cpu", kernelFunc: KH };
var YH = st(Ea, (e) => Math.cos(e));
var QH = { kernelName: Ea, backendName: "cpu", kernelFunc: YH };
var ZH = st(Ra, (e) => Math.cosh(e));
var JH = { kernelName: Ra, backendName: "cpu", kernelFunc: ZH };
function eq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { image: r, boxes: a, boxInd: i } = t, { cropSize: o, method: u, extrapolationValue: l } = s, [c, p, d, h] = r.shape, f = a.shape[0], [m, g] = o, b = Ae([f, m, g, h], "float32"), y = n.data.get(a.dataId).values, v = n.data.get(i.dataId).values, x = n.data.get(r.dataId).values, k = w.computeStrides(r.shape), I = w.computeStrides(b.shape);
  for (let $ = 0; $ < f; $++) {
    let R = $ * 4, E = y[R], P = y[R + 1], A = y[R + 2], O = y[R + 3], T = v[$];
    if (T >= c)
      continue;
    let z = m > 1 ? (A - E) * (p - 1) / (m - 1) : 0, W = g > 1 ? (O - P) * (d - 1) / (g - 1) : 0;
    for (let q = 0; q < m; q++) {
      let X = m > 1 ? E * (p - 1) + q * z : 0.5 * (E + A) * (p - 1);
      if (X < 0 || X > p - 1) {
        for (let Y = 0; Y < g; Y++)
          for (let Z = 0; Z < h; Z++) {
            let te = Z + Y * I[2] + q * I[1] + $ * I[0];
            b.values[te] = l;
          }
        continue;
      }
      if (u === "bilinear") {
        let Y = Math.floor(X), Z = Math.ceil(X), te = X - Y;
        for (let J = 0; J < g; J++) {
          let se = g > 1 ? P * (d - 1) + J * W : 0.5 * (P + O) * (d - 1);
          if (se < 0 || se > d - 1) {
            for (let de = 0; de < h; de++) {
              let me = de + J * I[2] + q * I[1] + $ * I[0];
              b.values[me] = l;
            }
            continue;
          }
          let ne = Math.floor(se), oe = Math.ceil(se), ae = se - ne;
          for (let de = 0; de < h; de++) {
            let me = de + ne * k[2] + Y * k[1] + T * k[0], ke = x[me];
            me = de + oe * k[2] + Y * k[1] + T * k[0];
            let Ie = x[me];
            me = de + ne * k[2] + Z * k[1] + T * k[0];
            let Re = x[me];
            me = de + oe * k[2] + Z * k[1] + T * k[0];
            let Pe = x[me], Xe = ke + (Ie - ke) * ae, Je = Re + (Pe - Re) * ae;
            me = de + J * I[2] + q * I[1] + $ * I[0], b.values[me] = Xe + (Je - Xe) * te;
          }
        }
      } else
        for (let Y = 0; Y < g; ++Y) {
          let Z = g > 1 ? P * (d - 1) + Y * W : 0.5 * (P + O) * (d - 1);
          if (Z < 0 || Z > d - 1) {
            for (let se = 0; se < h; se++) {
              let ne = se + Y * I[2] + q * I[1] + $ * I[0];
              b.values[ne] = l;
            }
            continue;
          }
          let te = Math.round(Z), J = Math.round(X);
          for (let se = 0; se < h; se++) {
            let ne = se + te * k[2] + J * k[1] + T * k[0], oe = se + Y * I[2] + q * I[1] + $ * I[0];
            b.values[oe] = x[ne];
          }
        }
    }
  }
  return n.makeTensorInfo(b.shape, b.dtype, b.values);
}
var tq = { kernelName: go, backendName: "cpu", kernelFunc: eq };
function nq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, exclusive: i, reverse: o } = s;
  be(r, "cumprod");
  let u = C.getAxesPermutation([a], r.shape.length), l = r;
  u != null && (l = wn({ inputs: { x: r }, backend: n, attrs: { perm: u } }));
  let c = C.getInnerMostAxes(1, r.shape.length)[0];
  if (c !== l.shape.length - 1)
    throw new Error(`backend.cumprod in CPU expects an inner-most axis=${l.shape.length - 1} but got axis=${c}`);
  let p = cn(l.dtype, "int32"), d = w.makeOnesTypedArray(w.sizeFromShape(l.shape), p), h = n.data.get(l.dataId).values, f = l.shape[l.shape.length - 1], m = o ? (b, y) => b + f - y - 1 : (b, y) => b + y;
  for (let b = 0; b < h.length; b += f)
    for (let y = 0; y < f; y++) {
      let v = m(b, y);
      if (y === 0)
        d[v] = i ? 1 : h[v];
      else {
        let x = m(b, y - 1);
        d[v] = i ? h[x] * d[x] : h[v] * d[x];
      }
    }
  let g = n.makeTensorInfo(l.shape, p, d);
  if (u != null) {
    let b = C.getUndoAxesPermutation(u), y = wn({ inputs: { x: g }, backend: n, attrs: { perm: b } });
    return n.disposeIntermediateTensorInfo(g), n.disposeIntermediateTensorInfo(l), y;
  }
  return g;
}
var sq = { kernelName: mo, backendName: "cpu", kernelFunc: nq };
function rq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, exclusive: i, reverse: o } = s;
  be(r, "cumsum");
  let u = C.getAxesPermutation([a], r.shape.length), l = r;
  u != null && (l = wn({ inputs: { x: r }, backend: n, attrs: { perm: u } }));
  let c = C.getInnerMostAxes(1, r.shape.length)[0];
  if (c !== l.shape.length - 1)
    throw new Error(`backend.cumsum in CPU expects an inner-most axis=${l.shape.length - 1} but got axis=${c}`);
  let p = cn(l.dtype, "int32"), d = w.makeZerosTypedArray(w.sizeFromShape(l.shape), p), h = n.data.get(l.dataId).values, f = l.shape[l.shape.length - 1], m = o ? (b, y) => b + f - y - 1 : (b, y) => b + y;
  for (let b = 0; b < h.length; b += f)
    for (let y = 0; y < f; y++) {
      let v = m(b, y);
      if (y === 0)
        d[v] = i ? 0 : h[v];
      else {
        let x = m(b, y - 1);
        d[v] = i ? h[x] + d[x] : h[v] + d[x];
      }
    }
  let g = n.makeTensorInfo(l.shape, p, d);
  if (u != null) {
    let b = C.getUndoAxesPermutation(u), y = wn({ inputs: { x: g }, backend: n, attrs: { perm: b } });
    return n.disposeIntermediateTensorInfo(g), n.disposeIntermediateTensorInfo(l), y;
  }
  return g;
}
var aq = { kernelName: Da, backendName: "cpu", kernelFunc: rq };
function iq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, weights: a } = t, { size: i, binaryOutput: o } = s;
  if (r.shape.length === 1) {
    let u = n.data.get(r.dataId).values, l = n.data.get(a.dataId).values, c = uv(u, l, a.dtype, a.shape, i);
    return n.makeTensorInfo([i], a.dtype, c);
  } else if (r.shape.length === 2) {
    let u = n.bufferSync(r), l = n.bufferSync(a), c = J0(u, l, i, o);
    return n.makeTensorInfo(c.shape, a.dtype, c.values);
  }
  throw new Error(`Error in denseBincount: input must be at most rank 2, but got rank${r.shape.length}.`);
}
var oq = { kernelName: vg, backendName: "cpu", kernelFunc: iq };
function uq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockSize: a, dataFormat: i } = s;
  w.assert(i === "NHWC", () => `Only NHWC dataFormat supported on CPU for depthToSpace. Got ${i}`);
  let o = r.shape[0], u = r.shape[1], l = r.shape[2], c = r.shape[3], p = u * a, d = l * a, h = c / (a * a), f = n.data.get(r.dataId).values, m = new Float32Array(o * p * d * h), g = 0;
  for (let b = 0; b < o; ++b)
    for (let y = 0; y < p; ++y) {
      let v = Math.floor(y / a), x = y % a;
      for (let k = 0; k < d; ++k) {
        let I = Math.floor(k / a), $ = k % a, R = (x * a + $) * h;
        for (let E = 0; E < h; ++E) {
          let A = E + R + c * (I + l * (v + u * b));
          m[g++] = f[A];
        }
      }
    }
  return n.makeTensorInfo([o, p, d, h], r.dtype, m);
}
var lq = { kernelName: bo, backendName: "cpu", kernelFunc: uq };
function GC(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dilations: u, dimRoundingMode: l } = s;
  be([r, a], "depthwiseConv2DNative");
  let c = w.computeStrides(r.shape), p = w.computeStrides(a.shape), d = u;
  d == null && (d = [1, 1]), w.assert(C.eitherStridesOrDilationsAreOne(i, d), () => `Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${i} and dilations '${d}'`);
  let h = C.computeConv2DInfo(r.shape, a.shape, i, d, o, l, true), { filterHeight: f, filterWidth: m, dilationHeight: g, dilationWidth: b, padInfo: y } = h, v = y.left, x = y.top, k = h.outChannels / h.inChannels, I = new Wt(h.outShape, r.dtype), $ = n.data.get(r.dataId).values, R = n.data.get(a.dataId).values, E = I.values;
  for (let P = 0; P < h.batchSize; ++P) {
    let A = P * c[0], O = P * I.strides[0];
    for (let T = 0; T < h.outHeight; ++T) {
      let z = O + T * I.strides[1], W = T * h.strideHeight - x;
      for (let q = 0; q < f; ++q) {
        let X = W + q * g;
        if (X < 0 || X >= h.inHeight)
          continue;
        let Y = q * p[0], Z = A + X * c[1];
        for (let te = 0; te < h.outWidth; ++te) {
          let J = z + te * I.strides[2], se = te * h.strideWidth - v;
          for (let ne = 0; ne < m; ++ne) {
            let oe = se + ne * b;
            if (oe < 0 || oe >= h.inWidth)
              continue;
            let ae = Y + ne * p[1], de = Z + oe * h.inChannels, me = J, ke = ae;
            for (let Ie = 0; Ie < h.inChannels; ++Ie) {
              let Re = $[de + Ie];
              for (let Pe = 0; Pe < k; ++Pe)
                E[me + Pe] += Re * R[ke + Pe];
              me += k, ke += k;
            }
          }
        }
      }
    }
  }
  return n.makeTensorInfo(I.shape, I.dtype, I.values);
}
var cq = { kernelName: Fa, backendName: "cpu", kernelFunc: GC };
function dq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, dy: a } = t, { strides: i, dilations: o, pad: u, dimRoundingMode: l, filterShape: c } = s;
  be([r, a], "depthwiseConv2dNativeBackpropFilter");
  let p = C.computeConv2DInfo(r.shape, c, i, o, u, l, true), { strideHeight: d, strideWidth: h, filterHeight: f, filterWidth: m } = p, g = new Wt(p.filterShape, "float32"), b = p.padInfo.left, y = p.padInfo.top, v = p.outChannels / p.inChannels, x = n.data.get(r.dataId).values, k = new Wt(r.shape, r.dtype, x), I = n.data.get(a.dataId).values, $ = new Wt(a.shape, a.dtype, I);
  for (let R = 0; R < f; ++R) {
    let E = Math.max(0, Math.ceil((y - R) / d)), P = Math.min(p.outHeight, (p.inHeight + y - R) / d);
    for (let A = 0; A < m; ++A) {
      let O = Math.max(0, Math.ceil((b - A) / h)), T = Math.min(p.outWidth, (p.inWidth + b - A) / h);
      for (let z = 0; z < p.outChannels; ++z) {
        let W = Math.trunc(z / v), q = z % v, X = 0;
        for (let Y = 0; Y < p.batchSize; ++Y)
          for (let Z = E; Z < P; ++Z) {
            let te = R + Z * d - y;
            for (let J = O; J < T; ++J) {
              let se = A + J * h - b;
              X += k.get(Y, te, se, W) * $.get(Y, Z, J, z);
            }
          }
        g.set(X, R, A, W, q);
      }
    }
  }
  return n.makeTensorInfo(g.shape, g.dtype, g.values);
}
var pq = { kernelName: xg, backendName: "cpu", kernelFunc: dq };
function hq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, filter: a } = t, { strides: i, dilations: o, pad: u, dimRoundingMode: l, inputShape: c } = s;
  be([r, a], "depthwiseConv2DNativeBackpropInput");
  let p = w.computeStrides(r.shape), d = w.computeStrides(a.shape), h = C.computeConv2DInfo(c, a.shape, i, o, u, l, true), f = new Wt(h.inShape, "float32"), m = f.values, [g, b, y] = f.strides, v = n.data.get(r.dataId).values, [x, k, I] = p, $ = n.data.get(a.dataId).values, [R, E, P] = d, { batchSize: A, filterHeight: O, filterWidth: T, inChannels: z, inHeight: W, inWidth: q, outChannels: X, outHeight: Y, outWidth: Z, strideHeight: te, strideWidth: J } = h, se = O - 1 - h.padInfo.top, ne = T - 1 - h.padInfo.left, oe = X / z;
  for (let ae = 0; ae < A; ++ae)
    for (let de = 0; de < z; ++de)
      for (let me = 0; me < W; ++me) {
        let ke = me - se, Ie = Math.max(0, Math.ceil(ke / te)), Re = Math.min(Y, (O + ke) / te);
        for (let Pe = 0; Pe < q; ++Pe) {
          let Xe = Pe - ne, Je = Math.max(0, Math.ceil(Xe / J)), Ye = Math.min(Z, (T + Xe) / J), tt = 0;
          for (let Ce = Ie; Ce < Re; ++Ce) {
            let ut = Ce * te - ke;
            for (let at = Je; at < Ye; ++at) {
              let Jt = at * J - Xe, Nt = x * ae + k * Ce + I * at, In = R * (O - 1 - ut) + E * (T - 1 - Jt) + P * de;
              for (let Rt = 0; Rt < oe; ++Rt) {
                let en = de * oe + Rt, Cn = v[Nt + en], Nn = $[In + Rt];
                tt += Cn * Nn;
              }
            }
          }
          m[g * ae + b * me + y * Pe + de] = tt;
        }
      }
  return n.makeTensorInfo(f.shape, f.dtype, f.values);
}
var fq = { kernelName: wg, backendName: "cpu", kernelFunc: hq };
function mq(e) {
  let { inputs: t, backend: n } = e, { x: s } = t, r = w.sizeFromShape(s.shape), a = n.data.get(s.dataId).values, i = Ae([r, r], s.dtype), o = i.values;
  for (let l = 0; l < a.length; l++)
    o[l * r + l] = a[l];
  let u = [...s.shape, ...s.shape];
  return n.makeTensorInfo(u, i.dtype, i.values);
}
var gq = { kernelName: kg, backendName: "cpu", kernelFunc: mq };
var bq = { kernelName: sp, backendName: "cpu", kernelFunc: ({ inputs: e, backend: t, attrs: n }) => {
  let { x: s, filter: r } = e, { strides: a, pad: i, dilations: o } = n, u = t, l = u.data.get(s.dataId).values, c = s.shape.length, p = u.data.get(r.dataId).values, d = r.shape.length, { batchSize: h, inHeight: f, inWidth: m, inChannels: g, outHeight: b, outWidth: y, padInfo: v, strideHeight: x, strideWidth: k, filterHeight: I, filterWidth: $, dilationHeight: R, dilationWidth: E, outShape: P } = C.computeDilation2DInfo(s.shape, r.shape, a, i, "NHWC", o), A = w.sizeFromShape(P), O = P.length, T = w.getArrayFromDType(s.dtype, A);
  for (let W = 0; W < h; ++W)
    for (let q = 0; q < b; ++q) {
      let X = q * x - v.top;
      for (let Y = 0; Y < y; ++Y) {
        let Z = Y * k - v.left;
        for (let te = 0; te < g; ++te) {
          let J = Number.MIN_SAFE_INTEGER;
          for (let ne = 0; ne < I; ++ne) {
            let oe = X + ne * R;
            if (oe >= 0 && oe < f)
              for (let ae = 0; ae < $; ++ae) {
                let de = Z + ae * E;
                if (de >= 0 && de < m) {
                  let me = w.locToIndex([W, oe, de, te], c, w.computeStrides(s.shape)), ke = w.locToIndex([ne, ae, te], d, w.computeStrides(r.shape)), Ie = l[me] + p[ke];
                  Ie > J && (J = Ie);
                }
              }
          }
          let se = w.locToIndex([W, q, Y, te], O, w.computeStrides(P));
          T[se] = J;
        }
      }
    }
  return { dataId: u.write(w.toTypedArray(T, s.dtype), P, s.dtype), shape: P, dtype: s.dtype };
} };
var yq = { kernelName: sm, backendName: "cpu", kernelFunc: ({ inputs: e, backend: t, attrs: n }) => {
  let { x: s, filter: r, dy: a } = e, { strides: i, pad: o, dilations: u } = n, l = t, c = w.toNestedArray(s.shape, l.data.get(s.dataId).values), p = w.toNestedArray(r.shape, l.data.get(r.dataId).values), { batchSize: d, inHeight: h, inWidth: f, inChannels: m, outHeight: g, outWidth: b, padInfo: y, strideHeight: v, strideWidth: x, filterHeight: k, filterWidth: I, dilationHeight: $, dilationWidth: R, outShape: E } = C.computeDilation2DInfo(s.shape, r.shape, i, o, "NHWC", u);
  w.assert(a.rank === E.length, () => `Error in ${sm}, dy must have the same rank as output ${E.length}, but got ${a.rank}`);
  let P = w.toNestedArray(E, l.data.get(a.dataId).values), A = w.makeZerosNestedTypedArray(r.shape, r.dtype);
  for (let T = 0; T < d; ++T)
    for (let z = 0; z < g; ++z) {
      let W = z * v - y.top;
      for (let q = 0; q < b; ++q) {
        let X = q * x - y.left;
        for (let Y = 0; Y < m; ++Y) {
          let Z = Number.MIN_SAFE_INTEGER, te = 0, J = 0;
          for (let se = 0; se < k; ++se) {
            let ne = W + se * $;
            if (ne >= 0 && ne < h)
              for (let oe = 0; oe < I; ++oe) {
                let ae = X + oe * R;
                if (ae >= 0 && ae < f) {
                  let de = c[T][ne][ae][Y] + p[se][oe][Y];
                  de > Z && (Z = de, te = se, J = oe);
                }
              }
          }
          A[te][J][Y] += P[T][z][q][Y];
        }
      }
    }
  return { dataId: l.write(w.toTypedArray(A, s.dtype), r.shape, r.dtype), shape: r.shape, dtype: r.dtype };
} };
var vq = { kernelName: nm, backendName: "cpu", kernelFunc: ({ inputs: e, backend: t, attrs: n }) => {
  let { x: s, filter: r, dy: a } = e, { strides: i, pad: o, dilations: u } = n, l = t, c = w.toNestedArray(s.shape, l.data.get(s.dataId).values), p = w.toNestedArray(r.shape, l.data.get(r.dataId).values), { batchSize: d, inHeight: h, inWidth: f, inChannels: m, outHeight: g, outWidth: b, padInfo: y, strideHeight: v, strideWidth: x, filterHeight: k, filterWidth: I, dilationHeight: $, dilationWidth: R, outShape: E } = C.computeDilation2DInfo(s.shape, r.shape, i, o, "NHWC", u);
  w.assert(a.rank === E.length, () => `Error in ${nm}, dy must have the same rank as output ${E.length}, but got ${a.rank}`);
  let P = w.toNestedArray(E, l.data.get(a.dataId).values), A = w.makeZerosNestedTypedArray(s.shape, s.dtype);
  for (let T = 0; T < d; ++T)
    for (let z = 0; z < g; ++z) {
      let W = z * v - y.top;
      for (let q = 0; q < b; ++q) {
        let X = q * x - y.left;
        for (let Y = 0; Y < m; ++Y) {
          let Z = Number.MIN_SAFE_INTEGER, te = W < 0 ? 0 : W, J = X < 0 ? 0 : X;
          for (let se = 0; se < k; ++se) {
            let ne = W + se * $;
            if (ne >= 0 && ne < h)
              for (let oe = 0; oe < I; ++oe) {
                let ae = X + oe * R;
                if (ae >= 0 && ae < f) {
                  let de = c[T][ne][ae][Y] + p[se][oe][Y];
                  de > Z && (Z = de, te = ne, J = ae);
                }
              }
          }
          A[T][te][J][Y] += P[T][z][q][Y];
        }
      }
    }
  return { dataId: l.write(w.toTypedArray(A, s.dtype), s.shape, s.dtype), shape: s.shape, dtype: s.dtype };
} };
function Ql(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  be(r, "sum");
  let o;
  r.dtype === "bool" ? o = xr({ inputs: { x: r }, backend: n, attrs: { dtype: "int32" } }) : o = Os({ inputs: { x: r }, backend: n });
  let u = o.shape.length, l = w.parseAxisParam(a, o.shape), c = C.getAxesPermutation(l, u), p = l, d = o;
  c != null && (d = wn({ inputs: { x: o }, backend: n, attrs: { perm: c } }), p = C.getInnerMostAxes(p.length, u)), C.assertAxesAreInnerMostDims("sum", p, d.shape.length);
  let [h, f] = C.computeOutAndReduceShapes(d.shape, p), m = C.upcastType(d.dtype, "int32"), g = Ld(n, h, m), b = w.sizeFromShape(f), y = n.data.get(g.dataId).values, v = n.data.get(d.dataId).values;
  for (let x = 0; x < y.length; ++x) {
    let k = x * b, I = 0;
    for (let $ = 0; $ < b; ++$)
      I += v[k + $];
    y[x] = I;
  }
  if (i) {
    let x = C.expandShapeToKeepDim(g.shape, l), k = g;
    g = pt({ inputs: { x: g }, backend: n, attrs: { shape: x } }), n.disposeIntermediateTensorInfo(k);
  }
  return n.disposeIntermediateTensorInfo(o), c != null && n.disposeIntermediateTensorInfo(d), g;
}
var xq = { kernelName: ci, backendName: "cpu", kernelFunc: Ql };
function wq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { equation: r } = s, a = t, { allDims: i, summedDims: o, idDims: u } = C.decodeEinsumEquation(r, a.length);
  C.checkEinsumDimSizes(i.length, u, a);
  let { path: l, steps: c } = C.getEinsumComputePath(o, u), p = c.length, d = null, h = i.length, f = [];
  for (let m = 0; m < p; ++m) {
    for (let g of c[m]) {
      let { permutationIndices: b, expandDims: y } = C.getEinsumPermutation(h, u[g]), v;
      C.isIdentityPermutation(b) ? v = a[g] : (v = wn({ inputs: { x: a[g] }, backend: n, attrs: { perm: b } }), f.push(v));
      let x = v.shape.slice();
      for (let k = 0; k < y.length; ++k)
        x.splice(y[k], 0, 1);
      w.arraysEqual(v.shape, x) || (v = pt({ inputs: { x: v }, backend: n, attrs: { shape: x } }), f.push(v)), d === null ? d = v : (d = Zp({ inputs: { a: v, b: d }, backend: n }), f.push(d));
    }
    m < p - 1 && (l[m] >= 0 && (d = Ql({ inputs: { x: d }, backend: n, attrs: { axis: l[m] - (i.length - h), keepDims: false } }), f.push(d)), h--);
  }
  for (let m of f)
    m !== d && n.disposeIntermediateTensorInfo(m);
  return d;
}
var kq = { kernelName: rp, backendName: "cpu", kernelFunc: wq };
function Sq(e) {
  let { inputs: t, backend: n } = e, { dy: s, y: r } = t;
  be([s, r], "eluGrad");
  let a = new Float32Array(w.sizeFromShape(r.shape)), i = n.data.get(r.dataId).values, o = n.data.get(s.dataId).values;
  for (let u = 0; u < i.length; ++u) {
    let l = i[u];
    l >= 1 ? a[u] = o[u] : a[u] = o[u] * (l + 1);
  }
  return n.makeTensorInfo(r.shape, "float32", a);
}
var Iq = { kernelName: Sg, backendName: "cpu", kernelFunc: Sq };
var Cq = C.ERF_P;
var Nq = C.ERF_A1;
var Tq = C.ERF_A2;
var $q = C.ERF_A3;
var _q = C.ERF_A4;
var Aq = C.ERF_A5;
var Eq = st(bl, (e) => {
  let t = Math.sign(e), n = Math.abs(e), s = 1 / (1 + Cq * n);
  return t * (1 - ((((Aq * s + _q) * s + $q) * s + Tq) * s + Nq) * s * Math.exp(-n * n));
});
var Rq = { kernelName: bl, backendName: "cpu", kernelFunc: Eq };
function Wd(e) {
  let { inputs: t, backend: n, attrs: s } = e, { input: r } = t, { dim: a } = s, i = r.shape.length, o = r.shape.slice(), u = a;
  return a < 0 && (w.assert(-(i + 1) <= a, () => `Axis must be in the interval [${-(i + 1)}, ${i}]`), u = i + a + 1), o.splice(u, 0, 1), pt({ inputs: { x: r }, backend: n, attrs: { shape: o } });
}
var Dq = { kernelName: vo, backendName: "cpu", kernelFunc: Wd };
var Fq = Et((e, t) => e / t);
var gv = Ht(Oa, Fq);
var Hm = { kernelName: Oa, backendName: "cpu", kernelFunc: gv };
function HC(e, t, n) {
  let s = e.shape, r = s[0], a = s[1], i = n.data.get(e.dataId), o = i.complexTensorInfos.real, u = i.complexTensorInfos.imag, l = [r, a], c = w.sizeFromShape(l), p = w.getTypedArrayFromDType("float32", c), d = w.getTypedArrayFromDType("float32", c);
  for (let g = 0; g < r; g++) {
    let b = ba({ inputs: { x: o }, backend: n, attrs: { begin: [g, 0], size: [1, a] } }), y = ba({ inputs: { x: u }, backend: n, attrs: { begin: [g, 0], size: [1, a] } }), v = En({ inputs: { real: b, imag: y }, backend: n }), { real: x, imag: k } = Oq(v, t, n), I = C.mergeRealAndImagArrays(x, k);
    for (let $ = 0; $ < a; $++) {
      let R = C.getComplexWithIndex(I, $);
      p[g * a + $] = R.real, d[g * a + $] = R.imag;
    }
    n.disposeIntermediateTensorInfo(b), n.disposeIntermediateTensorInfo(y), n.disposeIntermediateTensorInfo(v);
  }
  let h = n.makeTensorInfo(l, "float32", p), f = n.makeTensorInfo(l, "float32", d), m = En({ inputs: { real: h, imag: f }, backend: n });
  return n.disposeIntermediateTensorInfo(h), n.disposeIntermediateTensorInfo(f), m;
}
function Oq(e, t, n) {
  let s = w.sizeFromShape(e.shape), r = n.data.get(e.dataId), a = n.data.get(r.complexTensorInfos.real.dataId).values, i = n.data.get(r.complexTensorInfos.imag.dataId).values;
  if (Pq(s)) {
    let o = qm(a, i, s, t, n), u = [e.shape[0], e.shape[1]];
    if (t) {
      let l = n.makeTensorInfo(u, "float32", o.real), c = n.makeTensorInfo(u, "float32", o.imag), p = n.makeTensorInfo([], "float32", w.createScalarValue(s, "float32")), d = Os({ inputs: { x: p }, backend: n }), h = Hm.kernelFunc({ inputs: { a: l, b: p }, backend: n }), f = Hm.kernelFunc({ inputs: { a: c, b: d }, backend: n }), m = n.data.get(h.dataId).values, g = n.data.get(f.dataId).values;
      return n.disposeIntermediateTensorInfo(l), n.disposeIntermediateTensorInfo(c), n.disposeIntermediateTensorInfo(p), n.disposeIntermediateTensorInfo(d), n.disposeIntermediateTensorInfo(h), n.disposeIntermediateTensorInfo(f), { real: m, imag: g };
    }
    return o;
  } else {
    let o = C.mergeRealAndImagArrays(a, i), u = zq(o, s, t);
    return C.splitRealAndImagArrays(u);
  }
}
function Pq(e) {
  return (e & e - 1) === 0;
}
function qm(e, t, n, s, r) {
  if (n === 1)
    return { real: e, imag: t };
  let a = C.mergeRealAndImagArrays(e, t), i = n / 2, o = C.complexWithEvenIndex(a), u = o.real, l = o.imag, c = [u.length], p = r.makeTensorInfo(c, "float32", u), d = r.makeTensorInfo(c, "float32", l), h = En({ inputs: { real: p, imag: d }, backend: r }), f = C.complexWithOddIndex(a), m = f.real, g = f.imag, b = [m.length], y = r.makeTensorInfo(b, "float32", m), v = r.makeTensorInfo(b, "float32", g), x = En({ inputs: { real: y, imag: v }, backend: r }), k = qm(u, l, i, s, r), I = k.real, $ = k.imag, R = [I.length], E = r.makeTensorInfo(R, "float32", I), P = r.makeTensorInfo(R, "float32", $), A = En({ inputs: { real: E, imag: P }, backend: r }), O = qm(m, g, i, s, r), T = O.real, z = O.imag, W = [T.length], q = r.makeTensorInfo(W, "float32", T), X = r.makeTensorInfo(W, "float32", z), Y = En({ inputs: { real: q, imag: X }, backend: r }), Z = C.exponents(n, s), te = [Z.real.length], J = r.makeTensorInfo(te, "float32", Z.real), se = r.makeTensorInfo(te, "float32", Z.imag), ne = En({ inputs: { real: J, imag: se }, backend: r }), oe = Zp({ inputs: { a: ne, b: Y }, backend: r }), ae = io({ inputs: { a: A, b: oe }, backend: r }), de = fv({ inputs: { a: A, b: oe }, backend: r }), me = ga({ inputs: { input: ae }, backend: r }), ke = ga({ inputs: { input: de }, backend: r }), Ie = oo({ inputs: { input: ae }, backend: r }), Re = oo({ inputs: { input: de }, backend: r }), Pe = uo({ inputs: [me, ke], backend: r, attrs: { axis: 0 } }), Xe = uo({ inputs: [Ie, Re], backend: r, attrs: { axis: 0 } }), Je = r.data.get(Pe.dataId).values, Ye = r.data.get(Xe.dataId).values;
  return r.disposeIntermediateTensorInfo(p), r.disposeIntermediateTensorInfo(d), r.disposeIntermediateTensorInfo(h), r.disposeIntermediateTensorInfo(y), r.disposeIntermediateTensorInfo(v), r.disposeIntermediateTensorInfo(x), r.disposeIntermediateTensorInfo(E), r.disposeIntermediateTensorInfo(P), r.disposeIntermediateTensorInfo(A), r.disposeIntermediateTensorInfo(q), r.disposeIntermediateTensorInfo(X), r.disposeIntermediateTensorInfo(Y), r.disposeIntermediateTensorInfo(J), r.disposeIntermediateTensorInfo(se), r.disposeIntermediateTensorInfo(ne), r.disposeIntermediateTensorInfo(oe), r.disposeIntermediateTensorInfo(ae), r.disposeIntermediateTensorInfo(de), r.disposeIntermediateTensorInfo(me), r.disposeIntermediateTensorInfo(Ie), r.disposeIntermediateTensorInfo(ke), r.disposeIntermediateTensorInfo(Re), r.disposeIntermediateTensorInfo(Pe), r.disposeIntermediateTensorInfo(Xe), { real: Je, imag: Ye };
}
function zq(e, t, n) {
  let s = new Float32Array(t * 2);
  for (let r = 0; r < t; r++) {
    let a = 0, i = 0;
    for (let o = 0; o < t; o++) {
      let u = C.exponent(r * o, t, n), l = C.getComplexWithIndex(e, o);
      a += l.real * u.real - l.imag * u.imag, i += l.real * u.imag + l.imag * u.real;
    }
    n && (a /= t, i /= t), C.assignToTypedArray(s, a, i, r);
  }
  return s;
}
function Mq(e) {
  let { inputs: t, backend: n } = e, { input: s } = t, r = w.sizeFromShape(s.shape), a = s.shape[s.shape.length - 1], i = r / a, o = pt({ inputs: { x: s }, backend: n, attrs: { shape: [i, a] } }), u = HC(o, false, n), l = pt({ inputs: { x: u }, backend: n, attrs: { shape: s.shape } });
  return n.disposeIntermediateTensorInfo(o), n.disposeIntermediateTensorInfo(u), l;
}
var Lq = { kernelName: Ig, backendName: "cpu", kernelFunc: Mq };
function bv(e) {
  let { backend: t, attrs: n } = e, { shape: s, value: r, dtype: a } = n, i = a || w.inferDtype(r), o = w.getArrayFromDType(i, w.sizeFromShape(s));
  return Vq(o, r, i), t.makeTensorInfo(s, i, o);
}
var Bq = { kernelName: yl, backendName: "cpu", kernelFunc: bv };
function Vq(e, t, n) {
  e.fill(t);
}
var Wq = { kernelName: wo, backendName: "cpu", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { image: s } = e, r = n, a = w.getTypedArrayFromDType(s.dtype, w.sizeFromShape(s.shape)), [i, o, u, l] = s.shape, c = r.data.get(s.dataId).values;
  for (let d = 0; d < i; d++) {
    let h = d * u * o * l;
    for (let f = 0; f < o; f++) {
      let m = f * (u * l);
      for (let g = 0; g < u; g++) {
        let b = g * l;
        for (let y = 0; y < l; y++) {
          let v = Math.round(u - g - 1), x = h + m + b + y, k = c[x];
          if (v >= 0 && v < u) {
            let I = v * l, $ = h + m + I + y;
            k = c[$];
          }
          a[x] = k;
        }
      }
    }
  }
  return { dataId: r.write(a, s.shape, s.dtype), shape: s.shape, dtype: s.dtype };
} };
var Uq = Et((e, t) => Math.floor(e / t));
var Gq = Ht(La, Uq, null, "int32");
var Hq = { kernelName: La, backendName: "cpu", kernelFunc: Gq };
function qq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a, bias: i, preluActivationWeights: o } = t, { strides: u, pad: l, dataFormat: c, dilations: p, dimRoundingMode: d, activation: h, leakyreluAlpha: f } = s, m = UC({ inputs: { x: r, filter: a }, backend: n, attrs: { strides: u, pad: l, dataFormat: c, dilations: p, dimRoundingMode: d } });
  if (i) {
    let g = m;
    if (c === "NCHW" && i.shape.length === 1 && i.shape[0] !== 1) {
      let b = pt({ inputs: { x: i }, backend: n, attrs: { shape: [i.shape[0], 1, 1] } });
      m = io({ inputs: { a: m, b }, backend: n }), n.disposeIntermediateTensorInfo(b);
    } else
      m = io({ inputs: { a: m, b: i }, backend: n });
    n.disposeIntermediateTensorInfo(g);
  }
  if (h) {
    let g = m;
    if (c === "NCHW" && h === "prelu" && o.shape.length === 1 && o.shape[0] !== 1) {
      let b = pt({ inputs: { x: o }, backend: n, attrs: { shape: [o.shape[0], 1, 1] } });
      m = Vd(n, m, h, b, f), n.disposeIntermediateTensorInfo(b);
    } else
      m = Vd(n, m, h, o, f);
    n.disposeIntermediateTensorInfo(g);
  }
  return m;
}
var jq = { kernelName: ia, backendName: "cpu", kernelFunc: qq };
function Kq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a, bias: i, preluActivationWeights: o } = t, { strides: u, pad: l, dataFormat: c, dilations: p, dimRoundingMode: d, activation: h, leakyreluAlpha: f } = s, m = GC({ inputs: { x: r, filter: a }, backend: n, attrs: { strides: u, pad: l, dataFormat: c, dilations: p, dimRoundingMode: d } });
  if (i) {
    let g = m;
    m = io({ inputs: { a: m, b: i }, backend: n }), n.disposeIntermediateTensorInfo(g);
  }
  if (h) {
    let g = m;
    m = Vd(n, m, h, o, f), n.disposeIntermediateTensorInfo(g);
  }
  return m;
}
var Xq = { kernelName: oa, backendName: "cpu", kernelFunc: Kq };
function Yq(e) {
  let { inputs: t, backend: n } = e, { params: s, indices: r } = t, a = w.sizeFromShape(s.shape), i = r.shape, o = i[i.length - 1], [u, l, c, p] = C.prepareAndValidate(s, r);
  if (l === 0)
    return n.makeTensorInfo(u, s.dtype, []);
  let d = n.data.get(r.dataId).values, h = n.bufferSync(s), f = oC(d, h, s.dtype, l, o, c, p, s.shape, a);
  return n.makeTensorInfo(u, s.dtype, f.values);
}
var Qq = { kernelName: So, backendName: "cpu", kernelFunc: Yq };
function Zq(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, indices: a } = t, { axis: i, batchDims: o } = s;
  be([r, a], "gatherV2");
  let u = w.parseAxisParam(i, r.shape)[0], l = n.data.get(a.dataId).values, c = r.shape[u];
  for (let x = 0; x < l.length; ++x) {
    let k = l[x];
    w.assert(k <= c - 1 && k >= 0, () => `GatherV2: the index value ${k} is not in [0, ${c - 1}]`);
  }
  let p = o;
  o == null && (p = 0);
  let d = w.sizeFromShape(a.shape), h = C.segment_util.collectGatherOpShapeInfo(r, a, u, p), f = pt({ inputs: { x: r }, backend: n, attrs: { shape: [h.batchSize, h.outerSize, h.dimSize, h.sliceSize] } }), m = pt({ inputs: { x: a }, backend: n, attrs: { shape: [h.batchSize, d / h.batchSize] } }), g = [h.batchSize, h.outerSize, d / h.batchSize, h.sliceSize], b = n.bufferSync(m), y = n.bufferSync(f), v = uC(y, b, g);
  return n.disposeIntermediateTensorInfo(f), n.disposeIntermediateTensorInfo(m), n.makeTensorInfo(h.outputShape, v.dtype, v.values);
}
var Jq = { kernelName: ko, backendName: "cpu", kernelFunc: Zq };
function e6(e) {
  let { inputs: t, backend: n } = e, { input: s } = t, r = w.sizeFromShape(s.shape), a = s.shape[s.shape.length - 1], i = r / a, o = pt({ inputs: { x: s }, backend: n, attrs: { shape: [i, a] } }), u = HC(o, true, n), l = pt({ inputs: { x: u }, backend: n, attrs: { shape: s.shape } });
  return n.disposeIntermediateTensorInfo(o), n.disposeIntermediateTensorInfo(u), l;
}
var t6 = { kernelName: Cg, backendName: "cpu", kernelFunc: e6 };
var n6 = st(vl, (e) => Number.isFinite(e) ? 1 : 0, "bool");
var s6 = { kernelName: vl, backendName: "cpu", kernelFunc: n6 };
var r6 = st(xl, (e) => Math.abs(e) === 1 / 0 ? 1 : 0, "bool");
var a6 = { kernelName: xl, backendName: "cpu", kernelFunc: r6 };
var i6 = st(wl, (e) => Number.isNaN(e) ? 1 : 0, "bool");
var o6 = { kernelName: wl, backendName: "cpu", kernelFunc: i6 };
function u6(e) {
  let { backend: t, attrs: n } = e, { start: s, stop: r, num: a } = n, i = hC(s, r, a);
  return t.makeTensorInfo([i.length], "float32", i);
}
var l6 = { kernelName: Ng, backendName: "cpu", kernelFunc: u6 };
var c6 = st(kl, (e) => Math.log1p(e));
var d6 = { kernelName: kl, backendName: "cpu", kernelFunc: c6 };
var p6 = Et((e, t) => e && t);
var h6 = Ht(To, p6, null, "bool");
var f6 = { kernelName: To, backendName: "cpu", kernelFunc: h6 };
var m6 = st(Sl, (e) => e ? 0 : 1, "bool");
var g6 = { kernelName: Sl, backendName: "cpu", kernelFunc: m6 };
var b6 = Et((e, t) => e || t);
var y6 = Ht(ip, b6, null, "bool");
var v6 = { kernelName: ip, backendName: "cpu", kernelFunc: y6 };
function x6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { depthRadius: a, bias: i, alpha: o, beta: u } = s;
  be(r, "LRN");
  let l = r.shape[3], c = l - 1, p = n.data.get(r.dataId).values, d = w.sizeFromShape(r.shape), h = new Float32Array(d);
  function f(m) {
    let g = m % l, b = m - g + Math.max(0, g - a), y = m - g + Math.min(g + a, c), v = 0;
    for (; b <= y; b++) {
      let x = p[b];
      v += x * x;
    }
    return v;
  }
  for (let m = 0; m < d; m++) {
    let g = f(m), b = p[m] * Math.pow(i + o * g, -u);
    h[m] = b;
  }
  return n.makeTensorInfo(r.shape, r.dtype, h);
}
var w6 = { kernelName: op, backendName: "cpu", kernelFunc: x6 };
function k6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, y: a, dy: i } = t, { depthRadius: o, bias: u, alpha: l, beta: c } = s;
  be(i, "LRNGrad");
  let p = w.sizeFromShape(i.shape), d = i.shape[3], h = n.data.get(i.dataId).values, f = n.data.get(r.dataId).values, m = n.data.get(a.dataId).values, g = new Float32Array(p), b = p;
  for (let y = 0; y < b; y++) {
    let v = y % d, x = y - v + Math.max(0, v - o), k = y - v + Math.min(d, v + o + 1), I = 0;
    for (let $ = x; $ < k; $++)
      I += Math.pow(f[$], 2);
    I = l * I + u;
    for (let $ = x; $ < k; $++) {
      let R = -2 * l * c * f[$] * m[y] / I;
      y === $ && (R += Math.pow(I, -c)), R *= h[y], g[$] += R;
    }
  }
  return n.makeTensorInfo(i.shape, r.dtype, g);
}
var S6 = { kernelName: Tg, backendName: "cpu", kernelFunc: k6 };
function qC(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { reductionIndices: a, keepDims: i } = s, o = n, u = r.shape, l = u.length, c = w.parseAxisParam(a, u), p = c, d = C.getAxesPermutation(p, l), h = o.data.get(r.dataId).values;
  if (d != null) {
    let x = new Array(l);
    for (let k = 0; k < x.length; k++)
      x[k] = u[d[k]];
    h = dv(h, u, r.dtype, d, x), p = C.getInnerMostAxes(p.length, l), u = x;
  }
  be(r, "max"), C.assertAxesAreInnerMostDims("max", p, l);
  let [f, m] = C.computeOutAndReduceShapes(u, p), g = w.sizeFromShape(m), b = mC(h, g, f, r.dtype), y = o.write(b, f, r.dtype), v = f;
  return i && (v = C.expandShapeToKeepDim(f, c)), { dataId: y, shape: v, dtype: r.dtype };
}
var I6 = { kernelName: Ha, backendName: "cpu", kernelFunc: qC };
function C6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t;
  be(r, "maxPool");
  let { filterSize: a, strides: i, pad: o, dimRoundingMode: u } = s, l = 1;
  w.assert(C.eitherStridesOrDilationsAreOne(i, l), () => `Error in maxPool: Either strides or dilations must be 1. Got strides ${i} and dilations '${l}'`);
  let c = C.computePool2DInfo(r.shape, a, i, l, o, u), p;
  if (c.filterWidth === 1 && c.filterHeight === 1 && w.arraysEqual(c.inShape, c.outShape))
    p = Os({ inputs: { x: r }, backend: n });
  else {
    let d = n.data.get(r.dataId).values, h = w.computeStrides(r.shape), f = mv(d, r.shape, r.dtype, h, c, "max");
    p = n.makeTensorInfo(c.outShape, r.dtype, f.values);
  }
  return p;
}
var N6 = { kernelName: ja, backendName: "cpu", kernelFunc: C6 };
function T6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { filterSize: a, strides: i, pad: o, dimRoundingMode: u, dataFormat: l } = s;
  be(r, "maxPool3d");
  let c = C.computePool3DInfo(r.shape, a, i, 1, o, u, l), p = n.data.get(r.dataId).values, d = WC(p, r.shape, r.dtype, w.computeStrides(r.shape), c, "max");
  return n.makeTensorInfo(d.shape, "float32", d.values);
}
var $6 = { kernelName: up, backendName: "cpu", kernelFunc: T6 };
function _6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, input: a } = t, { filterSize: i, strides: o, pad: u, dimRoundingMode: l } = s;
  be([r, a], "maxPool3DGrad");
  let c = C.computePool3DInfo(a.shape, i, o, 1, u, l), p = n.bufferSync(a), d = gH(p, c), h = c.strideDepth, f = c.strideHeight, m = c.strideWidth, g = c.dilationDepth, b = c.dilationHeight, y = c.dilationWidth, v = c.effectiveFilterDepth, x = c.effectiveFilterHeight, k = c.effectiveFilterWidth, I = v - 1 - c.padInfo.front, $ = k - 1 - c.padInfo.left, R = x - 1 - c.padInfo.top, E = Ae(a.shape, "float32"), P = n.bufferSync(r);
  for (let A = 0; A < c.batchSize; ++A)
    for (let O = 0; O < c.inChannels; ++O)
      for (let T = 0; T < c.inDepth; ++T)
        for (let z = 0; z < c.inHeight; ++z)
          for (let W = 0; W < c.inWidth; ++W) {
            let q = T - I, X = z - R, Y = W - $, Z = 0;
            for (let te = 0; te < v; te += g) {
              let J = (q + te) / h;
              if (!(J < 0 || J >= c.outDepth || Math.floor(J) !== J))
                for (let se = 0; se < x; se += b) {
                  let ne = (X + se) / f;
                  if (!(ne < 0 || ne >= c.outHeight || Math.floor(ne) !== ne))
                    for (let oe = 0; oe < k; oe += y) {
                      let ae = (Y + oe) / m;
                      if (ae < 0 || ae >= c.outWidth || Math.floor(ae) !== ae)
                        continue;
                      let de = v * x * k - 1 - d.get(A, J, ne, ae, O), me = te * x * k + se * k + oe, ke = de === me ? 1 : 0;
                      if (ke === 0)
                        continue;
                      Z += P.get(A, J, ne, ae, O) * ke;
                    }
                }
            }
            E.set(Z, A, T, z, W, O);
          }
  return n.makeTensorInfo(E.shape, E.dtype, E.values);
}
var A6 = { kernelName: _g, backendName: "cpu", kernelFunc: _6 };
function E6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, input: a, output: i } = t, o = a;
  be([a, i], "maxPoolGrad");
  let { filterSize: u, strides: l, pad: c, dimRoundingMode: p } = s, d = C.computePool2DInfo(o.shape, u, l, 1, c, p), h = n.data.get(o.dataId).values, f = Ae(d.outShape, o.dtype, VC(h, o.shape, o.dtype, d).values), m = d.strideHeight, g = d.strideWidth, b = d.dilationHeight, y = d.dilationWidth, v = d.effectiveFilterHeight, x = d.effectiveFilterWidth, k = x - 1 - d.padInfo.left, I = v - 1 - d.padInfo.top, $ = Ae(o.shape, "float32"), R = n.data.get(r.dataId).values, E = Ae(r.shape, "float32", R);
  for (let P = 0; P < d.batchSize; ++P)
    for (let A = 0; A < d.inChannels; ++A)
      for (let O = 0; O < d.inHeight; ++O)
        for (let T = 0; T < d.inWidth; ++T) {
          let z = O - I, W = T - k, q = 0;
          for (let X = 0; X < v; X += b) {
            let Y = (z + X) / m;
            if (!(Y < 0 || Y >= d.outHeight || Math.floor(Y) !== Y))
              for (let Z = 0; Z < x; Z += y) {
                let te = (W + Z) / g;
                if (te < 0 || te >= d.outWidth || Math.floor(te) !== te)
                  continue;
                let J = v * x - 1 - f.get(P, Y, te, A), se = X * x + Z, ne = J === se ? 1 : 0;
                if (ne === 0)
                  continue;
                q += E.get(P, Y, te, A) * ne;
              }
          }
          $.set(q, P, O, T, A);
        }
  return n.makeTensorInfo($.shape, $.dtype, $.values);
}
var R6 = { kernelName: $g, backendName: "cpu", kernelFunc: E6 };
function D6(e, t, n, s, r) {
  let a = w.computeStrides(t), i = mv(e, t, n, a, r, "max"), o = VC(e, t, n, r, true, s);
  return [i.values, o.values];
}
var F6 = { kernelName: Ag, backendName: "cpu", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { x: s } = e, { filterSize: r, strides: a, pad: i, includeBatchInIndex: o } = t, u = n;
  be(s, "MaxPoolWithArgmax");
  let l = u.data.get(s.dataId).values, c = C.computePool2DInfo(s.shape, r, a, [1, 1], i), [p, d] = D6(l, s.shape, s.dtype, o, c), h = u.write(p, c.outShape, s.dtype), f = u.write(d, c.outShape, s.dtype);
  return [{ dataId: h, shape: c.outShape, dtype: s.dtype }, { dataId: f, shape: c.outShape, dtype: "int32" }];
} };
function O6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s, o = w.parseAxisParam(a, r.shape), l = C.computeOutAndReduceShapes(r.shape, o)[1], c = w.sizeFromShape(l), p = [], d = n.makeTensorInfo([], "float32", new Float32Array([c]));
  p.push(d);
  let h = xr({ inputs: { x: r }, backend: n, attrs: { dtype: "float32" } });
  p.push(h);
  let f = gv({ inputs: { a: h, b: d }, backend: n });
  p.push(f);
  let m = Ql({ inputs: { x: f }, backend: n, attrs: { axis: a, keepDims: i } });
  return p.forEach((g) => n.disposeIntermediateTensorInfo(g)), m;
}
var P6 = { kernelName: Ka, backendName: "cpu", kernelFunc: O6 };
function z6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  be(r, "min");
  let o = w.parseAxisParam(a, r.shape), u = o, l = C.getAxesPermutation(u, r.shape.length), c = r;
  l != null && (c = wn({ inputs: { x: r }, backend: n, attrs: { perm: l } }), u = C.getInnerMostAxes(u.length, r.shape.length)), C.assertAxesAreInnerMostDims("min", u, c.shape.length);
  let [p, d] = C.computeOutAndReduceShapes(c.shape, u), h = w.sizeFromShape(d), f = w.makeZerosTypedArray(w.sizeFromShape(p), c.dtype), m = n.data.get(c.dataId).values;
  for (let b = 0; b < f.length; ++b) {
    let y = b * h, v = m[y];
    for (let x = 0; x < h; ++x) {
      let k = m[y + x];
      (Number.isNaN(k) || k < v) && (v = k);
    }
    f[b] = v;
  }
  l != null && n.disposeIntermediateTensorInfo(c);
  let g = n.makeTensorInfo(p, c.dtype, f);
  if (i) {
    let b = C.expandShapeToKeepDim(p, o), y = pt({ inputs: { x: g }, backend: n, attrs: { shape: b } });
    return n.disposeIntermediateTensorInfo(g), y;
  }
  return g;
}
var M6 = { kernelName: Xa, backendName: "cpu", kernelFunc: z6 };
function L6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { paddings: a, mode: i } = s;
  be(r, "mirrorPad");
  let o = a.map((v, x) => v[0] + r.shape[x] + v[1]), u = a.map((v) => v[0]), l = a.map((v, x) => v[0] + r.shape[x]), c = i === "reflect" ? 0 : 1, p = n.data.get(r.dataId).values, d = r.shape.length, h = w.computeStrides(r.shape), f = w.sizeFromShape(o), m = o.length, g = w.computeStrides(o), b = w.getTypedArrayFromDType(r.dtype, f);
  for (let v = 0; v < f; v++) {
    let x = w.indexToLoc(v, m, g);
    for (let I = 0; I < m; I++)
      x[I] < u[I] ? x[I] = u[I] * 2 - x[I] - c : x[I] >= l[I] && (x[I] = (l[I] - 1) * 2 - x[I] + c);
    x = x.map((I, $) => I - u[$]);
    let k = w.locToIndex(x, d, h);
    b[v] = p[k];
  }
  return { dataId: n.write(b, o, r.dtype), shape: o, dtype: r.dtype };
}
var B6 = { kernelName: Qa, backendName: "cpu", kernelFunc: L6 };
var V6 = Et((e, t) => {
  let n = e % t;
  return e < 0 && t < 0 || e >= 0 && t >= 0 ? n : (n + t) % t;
});
var W6 = Ht(Il, V6);
var U6 = { kernelName: Il, backendName: "cpu", kernelFunc: W6 };
var G6 = wa(Xd());
function jC(e) {
  let { inputs: t, backend: n, attrs: s } = e, { logits: r } = t, { dim: a } = s, i = r.shape.length, o = a;
  if (o === -1 && (o = i - 1), o !== i - 1)
    throw Error(`Softmax along a non-last dimension is not yet supported. Logits was rank ${i} and dim was ${o}`);
  let u = w.parseAxisParam([o], r.shape), l = qC({ inputs: { x: r }, backend: n, attrs: { reductionIndices: u, keepDims: false } }), c = C.expandShapeToKeepDim(l.shape, u), p = pt({ inputs: { x: l }, backend: n, attrs: { shape: c } }), d = fv({ inputs: { a: r, b: p }, backend: n }), h = rC({ inputs: { x: d }, backend: n }), f = Ql({ inputs: { x: h }, backend: n, attrs: { axis: u, keepDims: false } }), m = pt({ inputs: { x: f }, backend: n, attrs: { shape: c } }), g = gv({ inputs: { a: h, b: m }, backend: n });
  return n.disposeIntermediateTensorInfo(l), n.disposeIntermediateTensorInfo(p), n.disposeIntermediateTensorInfo(d), n.disposeIntermediateTensorInfo(h), n.disposeIntermediateTensorInfo(f), n.disposeIntermediateTensorInfo(m), g;
}
var H6 = { kernelName: di, backendName: "cpu", kernelFunc: jC };
function q6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { logits: r } = t, { numSamples: a, seed: i, normalized: o } = s;
  be(r, "multinomial");
  let u = o ? r : jC({ inputs: { logits: r }, backend: n, attrs: { dim: -1 } }), l = u.shape[0], c = u.shape[1], p = n.data.get(u.dataId).values, d = [l, a], h = w.makeZerosTypedArray(w.sizeFromShape(d), "int32");
  for (let f = 0; f < l; ++f) {
    let m = f * c, g = new Float32Array(c - 1);
    g[0] = p[m];
    for (let v = 1; v < g.length; ++v)
      g[v] = g[v - 1] + p[m + v];
    let b = G6.alea(i.toString()), y = f * a;
    for (let v = 0; v < a; ++v) {
      let x = b();
      h[y + v] = g.length;
      for (let k = 0; k < g.length; k++)
        if (x < g[k]) {
          h[y + v] = k;
          break;
        }
    }
  }
  return o || n.disposeIntermediateTensorInfo(u), n.makeTensorInfo(d, "int32", h);
}
var j6 = { kernelName: Eg, backendName: "cpu", kernelFunc: q6 };
var K6 = ws.nonMaxSuppressionV3Impl;
function X6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { boxes: r, scores: a } = t, { maxOutputSize: i, iouThreshold: o, scoreThreshold: u } = s;
  be(r, "NonMaxSuppression");
  let l = n.data.get(r.dataId).values, c = n.data.get(a.dataId).values, { selectedIndices: p } = K6(l, c, i, o, u);
  return n.makeTensorInfo([p.length], "int32", new Int32Array(p));
}
var Y6 = { kernelName: Ao, backendName: "cpu", kernelFunc: X6 };
var Q6 = ws.nonMaxSuppressionV4Impl;
function Z6(e) {
  let { inputs: t, backend: n, attrs: s } = e, { boxes: r, scores: a } = t, { maxOutputSize: i, iouThreshold: o, scoreThreshold: u, padToMaxOutputSize: l } = s;
  be(r, "NonMaxSuppressionPadded");
  let c = n.data.get(r.dataId).values, p = n.data.get(a.dataId).values, { selectedIndices: d, validOutputs: h } = Q6(c, p, i, o, u, l);
  return [n.makeTensorInfo([d.length], "int32", new Int32Array(d)), n.makeTensorInfo([], "int32", new Int32Array([h]))];
}
var J6 = { kernelName: Cl, backendName: "cpu", kernelFunc: Z6 };
var ej = ws.nonMaxSuppressionV5Impl;
function tj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { boxes: r, scores: a } = t, { maxOutputSize: i, iouThreshold: o, scoreThreshold: u, softNmsSigma: l } = s;
  be(r, "NonMaxSuppressionWithScore");
  let c = n.data.get(r.dataId).values, p = n.data.get(a.dataId).values, d = i, h = o, f = u, m = l, { selectedIndices: g, selectedScores: b } = ej(c, p, d, h, f, m);
  return [n.makeTensorInfo([g.length], "int32", new Int32Array(g)), n.makeTensorInfo([b.length], "float32", new Float32Array(b))];
}
var nj = { kernelName: Eo, backendName: "cpu", kernelFunc: tj };
function sj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { indices: r } = t, { depth: a, onValue: i, offValue: o } = s;
  be(r, "oneHot");
  let u = w.sizeFromShape(r.shape), l = new Float32Array(u * a);
  l.fill(o);
  let c = n.data.get(r.dataId).values;
  for (let p = 0; p < u; ++p)
    c[p] >= 0 && c[p] < a && (l[p * a + c[p]] = i);
  return n.makeTensorInfo([...r.shape, a], "int32", l);
}
var rj = { kernelName: Do, backendName: "cpu", kernelFunc: sj };
function Ud(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (s.dtype === "string")
    throw new Error("zerosLike is not supported for string tensors");
  if (s.dtype === "complex64") {
    let r = ga({ inputs: { input: s }, backend: n }), a = Ud({ inputs: { x: r }, backend: n }), i = oo({ inputs: { input: s }, backend: n }), o = Ud({ inputs: { x: i }, backend: n }), u = En({ inputs: { real: a, imag: o }, backend: n });
    return n.disposeIntermediateTensorInfo(r), n.disposeIntermediateTensorInfo(a), n.disposeIntermediateTensorInfo(i), n.disposeIntermediateTensorInfo(o), u;
  } else
    return bv({ backend: n, attrs: { shape: s.shape, value: 0, dtype: s.dtype } });
}
var aj = { kernelName: Xo, backendName: "cpu", kernelFunc: Ud };
function KC(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (s.dtype === "string")
    throw new Error("onesLike is not supported for string tensors");
  if (s.dtype === "complex64") {
    let r = ga({ inputs: { input: s }, backend: n }), a = KC({ inputs: { x: r }, backend: n }), i = oo({ inputs: { input: s }, backend: n }), o = Ud({ inputs: { x: i }, backend: n }), u = En({ inputs: { real: a, imag: o }, backend: n });
    return n.disposeIntermediateTensorInfo(r), n.disposeIntermediateTensorInfo(a), n.disposeIntermediateTensorInfo(i), n.disposeIntermediateTensorInfo(o), u;
  } else
    return bv({ backend: n, attrs: { shape: s.shape, value: 1, dtype: s.dtype } });
}
var ij = { kernelName: Ro, backendName: "cpu", kernelFunc: KC };
function XC(e) {
  let { inputs: t, backend: n, attrs: s } = e, { axis: r } = s;
  if (t.length === 1)
    return Wd({ inputs: { input: t[0] }, backend: n, attrs: { dim: r } });
  let a = t[0].shape, i = t[0].dtype;
  t.forEach((c) => {
    w.assertShapesMatch(a, c.shape, "All tensors passed to stack must have matching shapes"), w.assert(i === c.dtype, () => "All tensors passed to stack must have matching dtypes");
  });
  let o = [], u = t.map((c) => {
    let p = Wd({ inputs: { input: c }, backend: n, attrs: { dim: r } });
    return o.push(p), p;
  }), l = uo({ inputs: u, backend: n, attrs: { axis: r } });
  return o.forEach((c) => n.disposeIntermediateTensorInfo(c)), l;
}
var oj = { kernelName: Fo, backendName: "cpu", kernelFunc: XC };
function uj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { paddings: a, constantValue: i } = s;
  be(r, "pad");
  let o = a.map((y, v) => y[0] + r.shape[v] + y[1]), u = a.map((y) => y[0]), l = n.data.get(r.dataId).values, c = w.sizeFromShape(r.shape), p = r.shape.length, d = w.computeStrides(r.shape), h = w.sizeFromShape(o), f = o.length, m = w.computeStrides(o), g = w.getTypedArrayFromDType(r.dtype, h);
  i !== 0 && g.fill(i);
  for (let y = 0; y < c; y++) {
    let x = w.indexToLoc(y, p, d).map((I, $) => I + u[$]), k = w.locToIndex(x, f, m);
    g[k] = l[y];
  }
  return { dataId: n.write(g, o, r.dtype), shape: o, dtype: r.dtype };
}
var YC = { kernelName: Ja, backendName: "cpu", kernelFunc: uj };
var lj = Et((e, t) => Math.pow(e, t));
var cj = Ht(ei, lj);
var dj = { kernelName: ei, backendName: "cpu", kernelFunc: cj };
function pj(e) {
  let { backend: t, attrs: n } = e, { start: s, stop: r, dtype: a, step: i } = n, o = pv(s, r, i, a);
  return t.makeTensorInfo([o.length], a, o);
}
var hj = { kernelName: Nl, backendName: "cpu", kernelFunc: pj };
var fj = st(Tl, (e) => 1 / e);
var mj = { kernelName: Tl, backendName: "cpu", kernelFunc: fj };
function gj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r } = t, { alignCorners: a, halfPixelCenters: i, size: o } = s;
  be(r, "resizeBilinear");
  let u = w.computeStrides(r.shape), [l, c] = o, [p, d, h, f] = r.shape, m = n.data.get(r.dataId).values, g = new Float32Array(w.sizeFromShape([p, l, c, f])), b = [a && l > 1 ? d - 1 : d, a && c > 1 ? h - 1 : h], y = [a && l > 1 ? l - 1 : l, a && c > 1 ? c - 1 : c], v = 0, x = b[0] / y[0], k = b[1] / y[1];
  for (let I = 0; I < p; I++)
    for (let $ = 0; $ < l; $++) {
      let R;
      i ? R = x * ($ + 0.5) - 0.5 : R = x * $;
      let E = Math.max(0, Math.floor(R)), P = R - E, A = Math.min(d - 1, Math.ceil(R)), O = I * u[0] + E * u[1], T = I * u[0] + A * u[1];
      for (let z = 0; z < c; z++) {
        let W;
        i ? W = k * (z + 0.5) - 0.5 : W = k * z;
        let q = Math.max(0, Math.floor(W)), X = W - q, Y = Math.min(h - 1, Math.ceil(W)), Z = O + q * u[2], te = T + q * u[2], J = O + Y * u[2], se = T + Y * u[2];
        for (let ne = 0; ne < f; ne++) {
          let oe = m[Z + ne], ae = m[te + ne], de = m[J + ne], me = m[se + ne], ke = oe + (de - oe) * X, Ie = ae + (me - ae) * X, Re = ke + (Ie - ke) * P;
          g[v++] = Re;
        }
      }
    }
  return n.makeTensorInfo([p, l, c, f], "float32", g);
}
var bj = { kernelName: ri, backendName: "cpu", kernelFunc: gj };
function yj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r, dy: a } = t, { alignCorners: i } = s;
  be([a, r], "resizeBilinearGrad");
  let o = w.computeStrides(r.shape), [u, l, c, p] = r.shape, [, d, h] = a.shape, f = new Float32Array(u * l * c * p), m = [i && d > 1 ? l - 1 : l, i && h > 1 ? c - 1 : c], g = [i && d > 1 ? d - 1 : d, i && h > 1 ? h - 1 : h], b = m[0] / g[0], y = m[1] / g[1], v = n.data.get(a.dataId).values, x = 0;
  for (let k = 0; k < u; k++) {
    let I = k * o[0];
    for (let $ = 0; $ < d; $++) {
      let R = $ * b, E = Math.floor(R), P = Math.min(Math.ceil(R), l - 1), A = I + E * o[1], O = I + P * o[1], T = R - E, z = 1 - T;
      for (let W = 0; W < h; W++) {
        let q = W * y, X = Math.floor(q), Y = Math.min(Math.ceil(q), c - 1), Z = q - X, te = 1 - Z, J = A + X * o[2], se = A + Y * o[2], ne = O + X * o[2], oe = O + Y * o[2], ae = z * te, de = z * Z, me = T * te, ke = T * Z;
        for (let Ie = 0; Ie < p; Ie++) {
          let Re = v[x++];
          f[J + Ie] += Re * ae, f[se + Ie] += Re * de, f[ne + Ie] += Re * me, f[oe + Ie] += Re * ke;
        }
      }
    }
  }
  return n.makeTensorInfo([u, c, l, p], "float32", f);
}
var vj = { kernelName: Dg, backendName: "cpu", kernelFunc: yj };
function xj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r } = t, { alignCorners: a, halfPixelCenters: i, size: o } = s;
  be(r, "resizeNearestNeighbor");
  let u = w.computeStrides(r.shape), [l, c] = o, [p, d, h, f] = r.shape, m = n.data.get(r.dataId).values, g = new Float32Array(p * l * c * f), b = [a && l > 1 ? d - 1 : d, a && c > 1 ? h - 1 : h], y = [a && l > 1 ? l - 1 : l, a && c > 1 ? c - 1 : c], v = b[0] / y[0], x = b[1] / y[1], k = 0;
  for (let I = 0; I < p; I++) {
    let $ = I * u[0];
    for (let R = 0; R < l; R++) {
      let E = i ? v * (R + 0.5) : v * R, P = Math.min(d - 1, a ? Math.round(E) : Math.floor(E));
      i && (P = Math.max(0, P));
      let A = $ + P * u[1];
      for (let O = 0; O < c; O++) {
        let T = i ? x * (O + 0.5) : x * O, z = Math.min(h - 1, a ? Math.round(T) : Math.floor(T));
        i && (z = Math.max(0, z));
        let W = A + z * u[2];
        for (let q = 0; q < f; q++) {
          let X = m[W + q];
          g[k++] = X;
        }
      }
    }
  }
  return n.makeTensorInfo([p, l, c, f], r.dtype, g);
}
var wj = { kernelName: $l, backendName: "cpu", kernelFunc: xj };
function kj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r, dy: a } = t, { alignCorners: i } = s;
  be([a, r], "resizeNearestNeighborGrad");
  let o = w.computeStrides(r.shape), u = w.computeStrides(a.shape), [l, c, p, d] = r.shape, [, h, f] = a.shape, m = new Float32Array(l * c * p * d), g = n.data.get(a.dataId).values, b = [i && h > 1 ? c - 1 : c, i && f > 1 ? p - 1 : p], y = [i && h > 1 ? h - 1 : h, i && f > 1 ? f - 1 : f], v = b[0] / y[0], x = b[1] / y[1], k = 1 / v, I = 1 / x, $ = Math.ceil(k) * 2 + 2, R = Math.ceil(I) * 2 + 2;
  for (let E = 0; E < l; E++) {
    let P = E * o[0];
    for (let A = 0; A < c; A++) {
      let O = P + A * o[1], T = Math.floor(A * k), z = Math.floor(T - $ / 2);
      for (let W = 0; W < p; W++) {
        let q = O + W * o[2], X = Math.floor(W * I), Y = Math.floor(X - R / 2);
        for (let Z = 0; Z < d; Z++) {
          let te = 0;
          for (let J = 0; J < $; J++) {
            let se = J + z;
            if (se < 0 || se >= h)
              continue;
            let ne = P + se * u[1], oe = se * v, ae = Math.min(c - 1, i ? Math.round(oe) : Math.floor(oe));
            if (A === ae)
              for (let de = 0; de < R; de++) {
                let me = de + Y;
                if (me < 0 || me >= f)
                  continue;
                let ke = ne + me * u[2], Ie = me * x, Re = Math.min(p - 1, i ? Math.round(Ie) : Math.floor(Ie));
                W === Re && (te += g[ke + Z]);
              }
          }
          m[q + Z] = te;
        }
      }
    }
  }
  return n.makeTensorInfo(r.shape, r.dtype, m);
}
var Sj = { kernelName: Rg, backendName: "cpu", kernelFunc: kj };
function Ij(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { dims: a } = s;
  be(r, "reverse");
  let i = r.shape.length, o = w.parseAxisParam(a, r.shape);
  if (i === 0)
    return Os({ inputs: { x: r }, backend: n });
  let u = new Wt(r.shape, r.dtype), l = n.bufferSync(r);
  for (let c = 0; c < u.size; c++) {
    let p = u.indexToLoc(c), d = p.slice();
    o.forEach((h) => d[h] = r.shape[h] - 1 - d[h]), u.set(l.get(...d), ...p);
  }
  return n.makeTensorInfo(u.shape, u.dtype, u.values);
}
var Cj = { kernelName: Po, backendName: "cpu", kernelFunc: Ij };
var Nj = { kernelName: Yo, backendName: "cpu", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { image: s } = e, { radians: r, fillValue: a, center: i } = t, o = n, u = w.getTypedArrayFromDType(s.dtype, w.sizeFromShape(s.shape)), [l, c, p, d] = s.shape, [h, f] = C.getImageCenter(i, c, p), m = 255, g = Math.sin(r), b = Math.cos(r), y = o.data.get(s.dataId).values;
  for (let x = 0; x < l; x++) {
    let k = x * p * c * d;
    for (let I = 0; I < c; I++) {
      let $ = I * (p * d);
      for (let R = 0; R < p; R++) {
        let E = R * d;
        for (let P = 0; P < d; P++) {
          let A = [l, I, R, P], O = A[2], T = A[1], z = (O - h) * b - (T - f) * g, W = (O - h) * g + (T - f) * b;
          z = Math.round(z + h), W = Math.round(W + f);
          let q = a;
          if (typeof a != "number" && (P === 3 ? q = m : q = a[P]), z >= 0 && z < p && W >= 0 && W < c) {
            let Y = W * (p * d), Z = z * d, te = k + Y + Z + P;
            q = y[te];
          }
          let X = k + $ + E + P;
          u[X] = q;
        }
      }
    }
  }
  return { dataId: o.write(u, s.shape, s.dtype), shape: s.shape, dtype: s.dtype };
} };
var Tj = st(zo, (e) => {
  let t = Math.floor(e);
  return e - t < 0.5 ? Math.floor(e) : e - t > 0.5 ? Math.ceil(e) : t % 2 === 0 ? t : t + 1;
});
var $j = { kernelName: zo, backendName: "cpu", kernelFunc: Tj };
function _j(e) {
  let { inputs: t, backend: n, attrs: s } = e, { indices: r, updates: a } = t, { shape: i } = s, { sliceRank: o, numUpdates: u, sliceSize: l, strides: c, outputSize: p } = C.calculateShapes(a, r, i), d = true, h = n.bufferSync(r), f = n.bufferSync(a), m = Ki(h, f, i, p, l, u, o, c, 0, d);
  return n.makeTensorInfo(i, m.dtype, m.values);
}
var Aj = { kernelName: Mo, backendName: "cpu", kernelFunc: _j };
function Ej(e, t) {
  let n = 0, s = e.length, r = 0;
  for (; n < s; )
    r = Math.floor((n + s) / 2), e[r] < t ? n = r + 1 : s = r;
  return s;
}
function Rj(e, t) {
  let n = 0, s = e.length, r = 0;
  for (; n < s; )
    r = Math.floor((n + s) / 2), e[r] <= t ? n = r + 1 : s = r;
  return s;
}
function Dj(e, t, n, s, r, a) {
  let i = w.getArrayFromDType("int32", n * r);
  for (let o = 0; o < n; ++o) {
    let u = e.slice(o * s, (o + 1) * s), l = o * r;
    for (let c = 0; c < r; ++c)
      i[l + c] = a === "left" ? Ej(u, t[c + l]) : Rj(u, t[c + l]);
  }
  return i;
}
function Fj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { sortedSequence: r, values: a } = t, { side: i } = s, o = n.data.get(r.dataId).values, u = n.data.get(a.dataId).values, l = Dj(o, u, r.shape[0], r.shape[1], a.shape[1], i);
  return n.makeTensorInfo(a.shape, "int32", l);
}
var Oj = { kernelName: Fg, backendName: "cpu", kernelFunc: Fj };
function Pj(e) {
  let { inputs: t, backend: n } = e, { condition: s, t: r, e: a } = t;
  be([s, r, a], "select");
  let i = s.shape.length, o = n.data.get(s.dataId).values, u = n.data.get(r.dataId).values, l = n.data.get(a.dataId).values, c = cn(r.dtype, a.dtype), p = w.makeZerosTypedArray(w.sizeFromShape(r.shape), c), d = 0, h = i === 0 || i > 1 || r.shape.length === 1 ? 1 : w.sizeFromShape(r.shape.slice(1));
  for (let f = 0; f < o.length; f++)
    for (let m = 0; m < h; m++)
      o[f] === 1 ? p[d++] = u[f] : p[d++] = l[f];
  return n.makeTensorInfo(r.shape, c, p);
}
var zj = { kernelName: Lo, backendName: "cpu", kernelFunc: Pj };
var Mj = C.SELU_SCALEALPHA;
var Lj = C.SELU_SCALE;
var Bj = st(_l, (e) => e >= 0 ? Lj * e : Mj * (Math.exp(e) - 1));
var Vj = { kernelName: _l, backendName: "cpu", kernelFunc: Bj };
var Wj = st(Al, (e) => e < 0 ? -1 : e > 0 ? 1 : 0);
var Uj = { kernelName: Al, backendName: "cpu", kernelFunc: Wj };
var Gj = st(oi, (e) => Math.sin(e));
var Hj = { kernelName: oi, backendName: "cpu", kernelFunc: Gj };
var qj = st(Vo, (e) => Math.sinh(e));
var jj = { kernelName: Vo, backendName: "cpu", kernelFunc: qj };
var Kj = 11920928955078125e-23;
var uw = Math.log(Kj) + 2;
var Xj = st(El, (e) => {
  let t = e > -uw, n = e < uw, s = Math.exp(e), r;
  return n ? r = s : t ? r = e : r = Math.log(1 + s), r;
});
var Yj = { kernelName: El, backendName: "cpu", kernelFunc: Xj };
function Qj(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockShape: a, paddings: i } = s;
  be([r], "spaceToBatchND");
  let o = w.sizeFromShape(a), u = [[0, 0]];
  u.push(...i);
  for (let I = 1 + a.length; I < r.shape.length; ++I)
    u.push([0, 0]);
  let l = YC.kernelFunc({ inputs: { x: r }, backend: n, attrs: { paddings: u, constantValue: 0 } }), c = C.getReshaped(l.shape, a, o, false), p = C.getPermuted(c.length, a.length, false), d = C.getReshapedPermuted(l.shape, a, o, false), m = pt({ inputs: { x: l }, backend: n, attrs: { shape: c } }), y = wn({ inputs: { x: m }, backend: n, attrs: { perm: p } }), k = pt({ inputs: { x: y }, backend: n, attrs: { shape: d } });
  return n.disposeIntermediateTensorInfo(l), n.disposeIntermediateTensorInfo(m), n.disposeIntermediateTensorInfo(y), k;
}
var Zj = { kernelName: Wo, backendName: "cpu", kernelFunc: Qj };
function Jj(e) {
  let { inputs: t, backend: n } = e, { indices: s, values: r, denseShape: a, defaultValue: i } = t;
  if (a.shape.length !== 1)
    throw new Error(`Dense shape must be a vector, saw:
        ${a.shape}`);
  if (s.shape.length !== 2)
    throw new Error(`Indices must be a matrix, saw:
        ${s.shape}`);
  if (r.shape.length !== 1)
    throw new Error(`Values must be a vector, saw:
        ${r.shape}`);
  if (i.shape.length !== 0)
    throw new Error(`Default value must be a scalar, saw:
        ${i.shape}`);
  let o = n.data.get(s.dataId).values, u = n.data.get(r.dataId).values, l = n.data.get(a.dataId).values, c = n.data.get(i.dataId).values[0], [p, d, h, f, m] = SC(o, s.shape, s.dtype, u, r.dtype, l, c);
  return [n.makeTensorInfo(d, s.dtype, p), n.makeTensorInfo([d[0]], r.dtype, h), n.makeTensorInfo([f.length], "bool", new Uint8Array(f.map((g) => Number(g)))), n.makeTensorInfo([m.length], s.dtype, new Int32Array(m))];
}
var e5 = { kernelName: cp, backendName: "cpu", kernelFunc: Jj };
function t5(e) {
  let { inputs: t, backend: n } = e, { inputIndices: s, inputShape: r, newShape: a } = t;
  if (s.shape.length !== 2)
    throw new Error(`Input indices should be a matrix but received shape
        ${s.shape}`);
  if (r.shape.length !== 1)
    throw new Error(`Input shape should be a vector but received shape
        ${r.shape}`);
  if (a.shape.length !== 1)
    throw new Error(`Target shape should be a vector but received shape ${a.shape}`);
  let i = Array.from(n.data.get(r.dataId).values), o = n.data.get(s.dataId).values, u = Array.from(n.data.get(a.dataId).values), [l, c, p] = IC(o, s.shape, s.dtype, i, u);
  return [n.makeTensorInfo(c, s.dtype, l), n.makeTensorInfo([p.length], a.dtype, new Int32Array(p))];
}
var n5 = { kernelName: Rl, backendName: "cpu", kernelFunc: t5 };
function s5(e) {
  let { inputs: t, backend: n } = e, { data: s, indices: r, segmentIds: a } = t;
  if (s.shape.length < 1)
    throw new Error("Data should be at least 1 dimensional but received scalar");
  if (r.shape.length !== 1)
    throw new Error(`Indices should be a vector but received shape
          ${r.shape}`);
  if (a.shape.length !== 1)
    throw new Error(`Segment ids should be a vector but received shape
          ${a.shape}`);
  if (r.shape[0] !== a.shape[0])
    throw new Error("segmentIds and indices should have same size.");
  let i = n.data.get(s.dataId).values, o = n.data.get(r.dataId).values, u = n.data.get(a.dataId).values, [l, c] = hv(i, s.shape, s.dtype, o, u, true);
  return n.makeTensorInfo(c, s.dtype, l);
}
var r5 = { kernelName: dp, backendName: "cpu", kernelFunc: s5 };
function a5(e) {
  let { inputs: t, backend: n } = e, { data: s, indices: r, segmentIds: a } = t;
  if (s.shape.length < 1)
    throw new Error("Data should be at least 1 dimensional but received scalar");
  if (r.shape.length !== 1)
    throw new Error(`Indices should be a vector but received shape
         ${r.shape}`);
  if (a.shape.length !== 1)
    throw new Error(`Segment ids should be a vector but received shape
         ${a.shape}`);
  if (r.shape[0] !== a.shape[0])
    throw new Error("segmentIds and indices should have same size.");
  let i = n.data.get(s.dataId).values, o = n.data.get(r.dataId).values, u = n.data.get(a.dataId).values, [l, c] = hv(i, s.shape, s.dtype, o, u);
  return n.makeTensorInfo(c, s.dtype, l);
}
var i5 = { kernelName: pp, backendName: "cpu", kernelFunc: a5 };
function o5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { sparseIndices: r, sparseValues: a, defaultValue: i } = t, { outputShape: o } = s, { sliceRank: u, numUpdates: l, sliceSize: c, strides: p, outputSize: d } = C.calculateShapes(a, r, o), h = false, f = n.bufferSync(r), m;
  switch (a.dtype) {
    case "bool": {
      let g = n.bufferSync(a), b = Boolean(n.data.get(i.dataId).values[0]);
      m = Ki(f, g, o, d, c, l, u, p, b, h);
      break;
    }
    case "float32": {
      let g = n.bufferSync(a), b = n.data.get(i.dataId).values[0];
      m = Ki(f, g, o, d, c, l, u, p, b, h);
      break;
    }
    case "int32": {
      let g = n.bufferSync(a), b = n.data.get(i.dataId).values[0];
      m = Ki(f, g, o, d, c, l, u, p, b, h);
      break;
    }
    case "string": {
      let g = n.bufferSync(a), b = w.decodeString(n.data.get(i.dataId).values[0]);
      m = Ki(f, g, o, d, c, l, u, p, b, h);
      break;
    }
    default:
      throw new Error(`Unsupported type ${a.dtype}`);
  }
  return n.makeTensorInfo(o, m.dtype, m.values);
}
var u5 = { kernelName: hp, backendName: "cpu", kernelFunc: o5 };
function l5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { numOrSizeSplits: a, axis: i } = s, o = w.parseAxisParam(i, r.shape)[0], u = C.prepareSplitSize(r, a, o), l = new Array(r.shape.length).fill(0), c = r.shape.slice();
  return u.map((p) => {
    let d = [...c];
    d[o] = p;
    let h = ba({ inputs: { x: r }, backend: n, attrs: { begin: l, size: d } });
    return l[o] += p, h;
  });
}
var c5 = { kernelName: Uo, backendName: "cpu", kernelFunc: l5 };
var d5 = { kernelName: Dl, backendName: "cpu", kernelFunc: ({ inputs: e, backend: t }) => {
  let { x: n } = e, s = t;
  be(n, "square");
  let r = s.data.get(n.dataId).values, a = new Float32Array(r.length);
  for (let o = 0; o < r.length; ++o) {
    let u = r[o];
    a[o] = u * u;
  }
  return { dataId: s.write(a, n.shape, n.dtype), shape: n.shape, dtype: n.dtype };
} };
var p5 = st(gi, (e, t) => {
  let n = t;
  return isNaN(e) ? NaN : e > 0 ? 1 : n.alpha;
});
var h5 = { kernelName: gi, backendName: "cpu", kernelFunc: p5 };
function f5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { begin: a, end: i, strides: o, beginMask: u, endMask: l, ellipsisMask: c, newAxisMask: p, shrinkAxisMask: d } = s;
  be(r, "stridedSlice");
  let { finalShapeSparse: h, finalShape: f, isIdentity: m, sliceDim0: g, isSimpleSlice: b, begin: y, end: v, strides: x } = wt.sliceInfo(r.shape, a, i, o, u, l, c, p, d), k;
  if (m)
    k = pt({ inputs: { x: r }, backend: n, attrs: { shape: f } });
  else if (g || b) {
    w.assert(r.shape.length >= 1, () => `Input must have rank at least 1, got: ${r.shape.length}`);
    let I = wt.computeOutShape(y, v, x), $ = ba({ inputs: { x: r }, backend: n, attrs: { begin: y, size: I } });
    k = pt({ inputs: { x: $ }, backend: n, attrs: { shape: f } }), n.disposeIntermediateTensorInfo($);
  } else {
    let I = n.bufferSync(r), $ = NC(h, I, x, y);
    k = n.makeTensorInfo(f, $.dtype, $.values);
  }
  return k;
}
var m5 = { kernelName: Go, backendName: "cpu", kernelFunc: f5 };
function g5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { separator: r, nGramWidths: a, leftPad: i, rightPad: o, padWidth: u, preserveShortSequences: l } = s, { data: c, dataSplits: p } = t, d = n.data.get(c.dataId).values, h = n.data.get(p.dataId).values, [f, m] = TC(d, h, r, a, i, o, u, l);
  return [n.makeTensorInfo([f.length], "string", f), n.makeTensorInfo(p.shape, "int32", m)];
}
var b5 = { kernelName: fp, backendName: "cpu", kernelFunc: g5 };
function y5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { skipEmpty: r } = s, { input: a, delimiter: i } = t;
  if (a.dtype !== "string")
    throw new Error("Input must be of datatype string");
  if (a.shape.length !== 1)
    throw new Error(`Input must be a vector, got shape: ${a.shape}`);
  if (i.shape.length !== 0)
    throw new Error(`Delimiter must be a scalar, got shape: ${i.shape}`);
  let o = n.data.get(a.dataId).values, u = n.data.get(i.dataId).values[0], [l, c, p] = $C(o, u, r), d = c.length;
  return [n.makeTensorInfo([d, 2], "int32", l), n.makeTensorInfo([d], "string", c), n.makeTensorInfo([2], "int32", new Int32Array(p))];
}
var v5 = { kernelName: Og, backendName: "cpu", kernelFunc: y5 };
function x5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { numBuckets: r } = s, { input: a } = t;
  if (a.dtype !== "string")
    throw new Error("Input must be of datatype string");
  if (r <= 0)
    throw new Error("Number of buckets must be at least 1");
  let i = n.data.get(a.dataId).values, o = _C(i, r);
  return n.makeTensorInfo(a.shape, "int32", o);
}
var w5 = { kernelName: Pg, backendName: "cpu", kernelFunc: x5 };
var k5 = st(Ho, (e) => Math.tan(e));
var S5 = { kernelName: Ho, backendName: "cpu", kernelFunc: k5 };
var I5 = st(fi, (e) => Math.tanh(e));
var C5 = { kernelName: fi, backendName: "cpu", kernelFunc: I5 };
function N5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { reps: a } = s;
  be(r, "tile");
  let i = EC(n.bufferSync(r), a);
  return n.makeTensorInfo(i.shape, i.dtype, i.values);
}
var T5 = { kernelName: Cr, backendName: "cpu", kernelFunc: N5 };
function $5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { k: a, sorted: i } = s;
  be(r, "topk");
  let o = n.data.get(r.dataId).values, [u, l] = DC(o, r.shape, r.dtype, a, i);
  return [n.makeTensorInfo(u.shape, u.dtype, u.values), n.makeTensorInfo(l.shape, l.dtype, l.values)];
}
var _5 = { kernelName: qo, backendName: "cpu", kernelFunc: $5 };
function A5(e) {
  let { inputs: t, attrs: n, backend: s } = e, { image: r, transforms: a } = t, { interpolation: i, fillMode: o, fillValue: u, outputShape: l } = n, [c, p, d, h] = r.shape, [f, m] = l != null ? l : [p, d], g = [c, f, m, h], b = w.computeStrides(r.shape), y = b[0], v = b[1], x = b[2], k = w.getTypedArrayFromDType(r.dtype, w.sizeFromShape(g));
  k.fill(u);
  let I = s.data.get(r.dataId).values, $ = s.data.get(a.dataId).values;
  for (let E = 0; E < c; ++E) {
    let P = a.shape[0] === 1 ? $ : $.subarray(E * 8, E * 8 + 8);
    for (let A = 0; A < f; ++A)
      for (let O = 0; O < m; ++O)
        for (let T = 0; T < h; ++T) {
          let z, W = P[6] * O + P[7] * A + 1;
          if (W === 0)
            continue;
          let q = (P[0] * O + P[1] * A + P[2]) / W, X = (P[3] * O + P[4] * A + P[5]) / W, Y = lw(q, d, o), Z = lw(X, p, o);
          switch (i) {
            case "nearest":
              z = P5(I, p, d, y, v, x, E, Z, Y, T, u);
              break;
            case "bilinear":
              z = z5(I, p, d, y, v, x, E, Z, Y, T, u);
              break;
            default:
              throw new Error(`Error in Transform: Expect 'nearest' or 'bilinear', but got ${i}`);
          }
          let te = E * y + A * v + O * x + T;
          k[te] = z;
        }
    return s.makeTensorInfo(g, r.dtype, k);
  }
  return { dataId: s.write(k, g, r.dtype), shape: r.shape, dtype: r.dtype };
}
var E5 = { kernelName: jo, backendName: "cpu", kernelFunc: A5 };
function lw(e, t, n) {
  switch (n) {
    case "reflect":
      return R5(e, t);
    case "wrap":
      return D5(e, t);
    case "nearest":
      return O5(e, t);
    case "constant":
    default:
      return F5(e, t);
  }
}
function R5(e, t) {
  let n = e;
  if (n < 0)
    if (t <= 1)
      n = 0;
    else {
      let s = 2 * t;
      n < s && (n = s * Math.trunc(-n / s) + n), n = n < -t ? n + s : -n - 1;
    }
  else if (n > t - 1)
    if (t <= 1)
      n = 0;
    else {
      let s = 2 * t;
      n -= s * Math.trunc(n / s), n >= t && (n = s - n - 1);
    }
  return w.clamp(0, n, t - 1);
}
function D5(e, t) {
  let n = e;
  if (n < 0)
    if (t <= 1)
      n = 0;
    else {
      let s = t - 1;
      n += t * (Math.trunc(-n / s) + 1);
    }
  else if (n > t - 1)
    if (t <= 1)
      n = 0;
    else {
      let s = t - 1;
      n -= t * Math.trunc(n / s);
    }
  return w.clamp(0, n, t - 1);
}
function F5(e, t) {
  return e;
}
function O5(e, t) {
  return w.clamp(0, e, t - 1);
}
function zu(e, t, n, s, r, a, i, o, u, l, c) {
  let p = i * s + o * r + u * a + l;
  return 0 <= o && o < t && 0 <= u && u < n ? e[p] : c;
}
function P5(e, t, n, s, r, a, i, o, u, l, c) {
  let p = Math.round(o), d = Math.round(u);
  return zu(e, t, n, s, r, a, i, p, d, l, c);
}
function z5(e, t, n, s, r, a, i, o, u, l, c) {
  let p = Math.floor(o), d = Math.floor(u), h = p + 1, f = d + 1, m = (f - u) * zu(e, t, n, s, r, a, i, p, d, l, c) + (u - d) * zu(e, t, n, s, r, a, i, p, f, l, c), g = (f - u) * zu(e, t, n, s, r, a, i, h, d, l, c) + (u - d) * zu(e, t, n, s, r, a, i, h, f, l, c);
  return (h - o) * m + (o - p) * g;
}
function M5(e) {
  let { inputs: t, attrs: n, backend: s } = e, { axis: r } = n, { x: a } = t;
  be(a, "unique");
  let i = s.data.get(a.dataId).values, { outputValues: o, outputShape: u, indices: l } = FC(i, r, a.shape, a.dtype);
  return [s.makeTensorInfo(u, a.dtype, o), s.makeTensorInfo([l.length], "int32", l)];
}
var L5 = { kernelName: zg, backendName: "cpu", kernelFunc: M5 };
function B5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { value: r } = t, { axis: a } = s;
  a < 0 && (a += r.shape.length);
  let i = r.shape.length, o = r.shape[a], u = new Array(i - 1), l = 0;
  for (let h = 0; h < i; h++)
    h !== a && (u[l++] = r.shape[h]);
  let c = new Array(i).fill(0), p = r.shape.slice();
  p[a] = 1;
  let d = new Array(o);
  for (let h = 0; h < d.length; h++) {
    c[a] = h;
    let f = ba({ inputs: { x: r }, backend: n, attrs: { begin: c, size: p } });
    d[h] = pt({ inputs: { x: f }, backend: n, attrs: { shape: u } }), n.disposeIntermediateTensorInfo(f);
  }
  return d;
}
var V5 = { kernelName: Ko, backendName: "cpu", kernelFunc: B5 };
function W5(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, segmentIds: a } = t, { numSegments: i } = s;
  be(r, "unsortedSegmentSum");
  let o = r.shape.length, u = a.shape.length, l = [], c = [], p = o - u, d = a;
  for (let f = 0; f < p; ++f) {
    let m = Wd({ inputs: { input: d }, backend: n, attrs: { dim: f + 1 } });
    d = m, c.push(m);
  }
  for (let f = 0; f < i; ++f) {
    let m = w.createScalarValue(f, "int32"), g = n.makeTensorInfo([], "int32", m), b = nC({ inputs: { a: g, b: d }, backend: n }), y = xr({ inputs: { x: b }, backend: n, attrs: { dtype: "float32" } }), v = Zp({ inputs: { a: y, b: r }, backend: n }), x = Ql({ inputs: { x: v }, backend: n, attrs: { axis: 0, keepDims: false } });
    l.push(x), c.push(g), c.push(b), c.push(y), c.push(v), c.push(x);
  }
  let h = XC({ inputs: l, backend: n, attrs: { axis: 0 } });
  return c.forEach((f) => n.disposeIntermediateTensorInfo(f)), h;
}
var U5 = { kernelName: mp, backendName: "cpu", kernelFunc: W5 };
var G5 = [GG, zU, qG, KG, UU, YG, ZG, eH, nH, rH, iH, uH, cH, hH, mH, yH, xH, kH, IH, WG, NH, $H, AH, RH, VU, HU, FH, MU, PH, MH, LH, VH, UH, HH, jH, XH, QH, JH, tq, sq, aq, oq, lq, cq, pq, fq, gq, bq, yq, vq, kq, OG, Iq, qU, Rq, jU, Dq, XU, Lq, Bq, Wq, QU, Hq, jq, Xq, Qq, Jq, JU, tG, LU, t6, zH, s6, a6, o6, PG, sG, aG, l6, oG, d6, f6, g6, v6, w6, S6, I6, lG, N6, $6, A6, R6, F6, P6, M6, dG, B6, U6, j6, hG, mG, Y6, J6, nj, bG, rj, ij, oj, YC, dj, MG, xG, hj, BU, Hm, mj, LG, BG, VG, bj, vj, wj, Sj, Cj, Nj, $j, kG, Aj, Oj, zj, Vj, IG, Uj, Hj, jj, CG, H6, Yj, Zj, e5, n5, r5, i5, u5, c5, $G, d5, AG, h5, m5, b5, v5, w5, FG, xq, S5, C5, T5, _5, E5, yG, L5, V5, U5, aj];
for (let e of G5)
  Fl(e);
var H5 = {};
Ee(H5, { assertNotComplex: () => iu, bindCanvasToFramebuffer: () => sK, bindColorTextureToFramebuffer: () => ud, bindTextureToProgramUniformSampler: () => d1, bindTextureUnit: () => u1, bindVertexBufferToProgramAttribute: () => jm, callAndCheck: () => fe, canBeRepresented: () => QC, createFragmentShader: () => e1, createFramebuffer: () => o1, createProgram: () => t1, createStaticIndexBuffer: () => r1, createStaticVertexBuffer: () => s1, createTexture: () => a1, createVertexShader: () => JC, getBatchDim: () => ya, getExtensionOrThrow: () => Mu, getFramebufferErrorMessage: () => p1, getMaxTexturesInShader: () => g1, getNumChannels: () => tK, getProgramUniformLocation: () => c1, getProgramUniformLocationOrThrow: () => l1, getRowsCols: () => va, getShapeAs3D: () => ld, getTextureShapeFromLogicalShape: () => f1, getWebGLDisjointQueryTimerVersion: () => b1, getWebGLErrorMessage: () => ZC, getWebGLMaxTextureSize: () => m1, hasExtension: () => Ln, isCapableOfRenderingToFloatTexture: () => y1, isDownloadFloatTextureEnabled: () => v1, isReshapeFree: () => rl, isWebGLFenceEnabled: () => x1, isWebGLVersionEnabled: () => Xm, linkProgram: () => n1, logShaderSourceAndInfoLog: () => vv, resetMaxTextureSize: () => rK, resetMaxTexturesInShader: () => aK, unbindColorTextureFromFramebuffer: () => Km, unbindTextureUnit: () => nK, validateFramebuffer: () => Lu, validateProgram: () => od, validateTextureSize: () => i1 });
var Yr = {};
var Jf = { alpha: false, antialias: false, premultipliedAlpha: false, preserveDrawingBuffer: false, depth: false, stencil: false, failIfMajorPerformanceCaveat: true };
function q5(e, t) {
  Yr[e] = t;
}
function xs(e, t) {
  if (!(e in Yr) || t != null) {
    let s = K5(e, t);
    if (s !== null)
      Yr[e] = s;
    else
      return console.log("Could not get context for WebGL version", e), null;
  }
  let n = Yr[e];
  return n == null || n.isContextLost() ? (delete Yr[e], xs(e)) : (n.disable(n.DEPTH_TEST), n.disable(n.STENCIL_TEST), n.disable(n.BLEND), n.disable(n.DITHER), n.disable(n.POLYGON_OFFSET_FILL), n.disable(n.SAMPLE_COVERAGE), n.enable(n.SCISSOR_TEST), n.enable(n.CULL_FACE), n.cullFace(n.BACK), Yr[e]);
}
function j5(e) {
  if (typeof OffscreenCanvas != "undefined" && e === 2)
    return new OffscreenCanvas(300, 150);
  if (typeof document != "undefined")
    return document.createElement("canvas");
  throw new Error("Cannot create a canvas in this context");
}
function K5(e, t) {
  if (e !== 1 && e !== 2)
    throw new Error("Cannot get WebGL rendering context, WebGL is disabled.");
  let n = t == null ? j5(e) : t;
  return n.addEventListener("webglcontextlost", (s) => {
    s.preventDefault(), delete Yr[e];
  }, false), e === 1 ? n.getContext("webgl", Jf) || n.getContext("experimental-webgl", Jf) : n.getContext("webgl2", Jf);
}
function Zl(e, t) {
  return [t, e];
}
function X5(e, t) {
  return e * t;
}
function Jc(e) {
  let t = w.sizeFromShape(e), n = Math.ceil(t / 4);
  return w.sizeToSquarishShape(n);
}
function au(e, t) {
  return [Math.max(1, Math.ceil(t / 2)), Math.max(1, Math.ceil(e / 2))];
}
function Y5(e, t) {
  let [n, s] = au(e, t);
  return n * s * 4;
}
function yv(e, t) {
  let n = e, s, r, a, i, o, u, l, c, p, d;
  return K().getNumber("WEBGL_VERSION") === 2 ? (s = n.R32F, r = n.R16F, a = n.RGBA16F, i = n.RGBA32F, o = n.RED, l = 4, c = 1, p = n.HALF_FLOAT, d = n.FLOAT, u = n.RGBA8) : (s = e.RGBA, r = e.RGBA, a = e.RGBA, i = n.RGBA, o = e.RGBA, l = 4, c = 4, p = t != null ? t.HALF_FLOAT_OES : null, d = e.FLOAT, u = e.RGBA), { internalFormatFloat: s, internalFormatHalfFloat: r, internalFormatPackedHalfFloat: a, internalFormatPackedFloat: i, textureFormatFloat: o, downloadTextureFormat: u, downloadUnpackNumChannels: l, defaultNumChannels: c, textureTypeHalfFloat: p, textureTypeFloat: d };
}
function fe(e, t) {
  let n = t();
  return K().getBool("DEBUG") && Q5(e), n;
}
function Q5(e) {
  let t = e.getError();
  if (t !== e.NO_ERROR)
    throw new Error("WebGL Error: " + ZC(e, t));
}
var Z5 = 596e-10;
var J5 = 65504;
function QC(e) {
  return !!(K().getBool("WEBGL_RENDER_FLOAT32_ENABLED") || e === 0 || Z5 < Math.abs(e) && Math.abs(e) < J5);
}
function ZC(e, t) {
  switch (t) {
    case e.NO_ERROR:
      return "NO_ERROR";
    case e.INVALID_ENUM:
      return "INVALID_ENUM";
    case e.INVALID_VALUE:
      return "INVALID_VALUE";
    case e.INVALID_OPERATION:
      return "INVALID_OPERATION";
    case e.INVALID_FRAMEBUFFER_OPERATION:
      return "INVALID_FRAMEBUFFER_OPERATION";
    case e.OUT_OF_MEMORY:
      return "OUT_OF_MEMORY";
    case e.CONTEXT_LOST_WEBGL:
      return "CONTEXT_LOST_WEBGL";
    default:
      return `Unknown error code ${t}`;
  }
}
function Mu(e, t) {
  return Qs(e, () => e.getExtension(t), 'Extension "' + t + '" not supported on this browser.');
}
function JC(e, t) {
  let n = Qs(e, () => e.createShader(e.VERTEX_SHADER), "Unable to create vertex WebGLShader.");
  if (fe(e, () => e.shaderSource(n, t)), fe(e, () => e.compileShader(n)), e.getShaderParameter(n, e.COMPILE_STATUS) === false)
    throw console.log(e.getShaderInfoLog(n)), new Error("Failed to compile vertex shader.");
  return n;
}
function e1(e, t) {
  let n = Qs(e, () => e.createShader(e.FRAGMENT_SHADER), "Unable to create fragment WebGLShader.");
  if (fe(e, () => e.shaderSource(n, t)), fe(e, () => e.compileShader(n)), K().get("ENGINE_COMPILE_ONLY"))
    return n;
  if (e.getShaderParameter(n, e.COMPILE_STATUS) === false)
    throw vv(t, e.getShaderInfoLog(n)), new Error("Failed to compile fragment shader.");
  return n;
}
var eK = /ERROR: [0-9]+:([0-9]+):/g;
function vv(e, t) {
  let n = eK.exec(t);
  if (n == null) {
    console.log(`Couldn't parse line number in error: ${t}`), console.log(e);
    return;
  }
  let s = +n[1], r = e.split(`
`), a = r.length.toString().length + 2, i = r.map((p, d) => w.rightPad((d + 1).toString(), a) + p), o = 0;
  for (let p = 0; p < i.length; p++)
    o = Math.max(i[p].length, o);
  let u = i.slice(0, s - 1), l = i.slice(s - 1, s), c = i.slice(s);
  console.log(u.join(`
`)), console.log(t.split(`
`)[0]), console.log(`%c ${w.rightPad(l[0], o)}`, "border:1px solid red; background-color:#e3d2d2; color:#a61717"), console.log(c.join(`
`));
}
function t1(e) {
  return Qs(e, () => e.createProgram(), "Unable to create WebGLProgram.");
}
function n1(e, t) {
  if (fe(e, () => e.linkProgram(t)), !K().get("ENGINE_COMPILE_ONLY") && e.getProgramParameter(t, e.LINK_STATUS) === false)
    throw console.log(e.getProgramInfoLog(t)), new Error("Failed to link vertex and fragment shaders.");
}
function od(e, t) {
  if (fe(e, () => e.validateProgram(t)), e.getProgramParameter(t, e.VALIDATE_STATUS) === false)
    throw console.log(e.getProgramInfoLog(t)), new Error("Shader program validation failed.");
}
function s1(e, t) {
  let n = Qs(e, () => e.createBuffer(), "Unable to create WebGLBuffer");
  return fe(e, () => e.bindBuffer(e.ARRAY_BUFFER, n)), fe(e, () => e.bufferData(e.ARRAY_BUFFER, t, e.STATIC_DRAW)), n;
}
function r1(e, t) {
  let n = Qs(e, () => e.createBuffer(), "Unable to create WebGLBuffer");
  return fe(e, () => e.bindBuffer(e.ELEMENT_ARRAY_BUFFER, n)), fe(e, () => e.bufferData(e.ELEMENT_ARRAY_BUFFER, t, e.STATIC_DRAW)), n;
}
function tK() {
  return K().getNumber("WEBGL_VERSION") === 2 ? 1 : 4;
}
function a1(e) {
  return Qs(e, () => e.createTexture(), "Unable to create WebGLTexture.");
}
function i1(e, t) {
  let n = K().getNumber("WEBGL_MAX_TEXTURE_SIZE");
  if (e <= 0 || t <= 0) {
    let s = `[${e}x${t}]`;
    throw new Error("Requested texture size " + s + " is invalid.");
  }
  if (e > n || t > n) {
    let s = `[${e}x${t}]`, r = `[${n}x${n}]`;
    throw new Error("Requested texture size " + s + " greater than WebGL maximum on this browser / GPU " + r + ".");
  }
}
function o1(e) {
  return Qs(e, () => e.createFramebuffer(), "Unable to create WebGLFramebuffer.");
}
function jm(e, t, n, s, r, a, i) {
  let o = e.getAttribLocation(t, n);
  return o === -1 ? false : (fe(e, () => e.bindBuffer(e.ARRAY_BUFFER, s)), fe(e, () => e.vertexAttribPointer(o, r, e.FLOAT, false, a, i)), fe(e, () => e.enableVertexAttribArray(o)), true);
}
function u1(e, t, n) {
  h1(e, n), fe(e, () => e.activeTexture(e.TEXTURE0 + n)), fe(e, () => e.bindTexture(e.TEXTURE_2D, t));
}
function nK(e, t) {
  h1(e, t), fe(e, () => e.activeTexture(e.TEXTURE0 + t)), fe(e, () => e.bindTexture(e.TEXTURE_2D, null));
}
function l1(e, t, n) {
  return Qs(e, () => e.getUniformLocation(t, n), 'uniform "' + n + '" not present in program.');
}
function c1(e, t, n) {
  return e.getUniformLocation(t, n);
}
function d1(e, t, n, s) {
  fe(e, () => u1(e, t, s)), fe(e, () => e.uniform1i(n, s));
}
function sK(e) {
  fe(e, () => e.bindFramebuffer(e.FRAMEBUFFER, null)), fe(e, () => e.viewport(0, 0, e.canvas.width, e.canvas.height)), fe(e, () => e.scissor(0, 0, e.canvas.width, e.canvas.height));
}
function ud(e, t, n) {
  fe(e, () => e.bindFramebuffer(e.FRAMEBUFFER, n)), fe(e, () => e.framebufferTexture2D(e.FRAMEBUFFER, e.COLOR_ATTACHMENT0, e.TEXTURE_2D, t, 0));
}
function Km(e, t) {
  fe(e, () => e.bindFramebuffer(e.FRAMEBUFFER, t)), fe(e, () => e.framebufferTexture2D(e.FRAMEBUFFER, e.COLOR_ATTACHMENT0, e.TEXTURE_2D, null, 0));
}
function Lu(e) {
  let t = e.checkFramebufferStatus(e.FRAMEBUFFER);
  if (t !== e.FRAMEBUFFER_COMPLETE)
    throw new Error("Error binding framebuffer: " + p1(e, t));
}
function p1(e, t) {
  switch (t) {
    case e.FRAMEBUFFER_INCOMPLETE_ATTACHMENT:
      return "FRAMEBUFFER_INCOMPLETE_ATTACHMENT";
    case e.FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT:
      return "FRAMEBUFFER_INCOMPLETE_MISSING_ATTACHMENT";
    case e.FRAMEBUFFER_INCOMPLETE_DIMENSIONS:
      return "FRAMEBUFFER_INCOMPLETE_DIMENSIONS";
    case e.FRAMEBUFFER_UNSUPPORTED:
      return "FRAMEBUFFER_UNSUPPORTED";
    default:
      return `unknown error ${t}`;
  }
}
function Qs(e, t, n) {
  let s = fe(e, () => t());
  if (s == null)
    throw new Error(n);
  return s;
}
function h1(e, t) {
  let n = e.MAX_COMBINED_TEXTURE_IMAGE_UNITS - 1, s = t + e.TEXTURE0;
  if (s < e.TEXTURE0 || s > n) {
    let r = `[gl.TEXTURE0, gl.TEXTURE${n}]`;
    throw new Error(`textureUnit must be in ${r}.`);
  }
}
function ya(e, t = 2) {
  return w.sizeFromShape(e.slice(0, e.length - t));
}
function va(e) {
  if (e.length === 0)
    throw Error("Cannot get rows and columns of an empty shape array.");
  return [e.length > 1 ? e[e.length - 2] : 1, e[e.length - 1]];
}
function ld(e) {
  let t = [1, 1, 1];
  return e.length === 0 || e.length === 1 && e[0] === 1 || (t = [ya(e), ...va(e)]), t;
}
function f1(e, t = false) {
  let n = K().getNumber("WEBGL_MAX_TEXTURE_SIZE");
  t && (n = n * 2, e = e.map((r, a) => a >= e.length - 2 ? w.nearestLargerEven(e[a]) : e[a]), e.length === 1 && (e = [2, e[0]])), e.length !== 2 && (e = w.squeezeShape(e).newShape);
  let s = w.sizeFromShape(e);
  if (e.length <= 1 && s <= n)
    return [1, s];
  if (e.length === 2 && e[0] <= n && e[1] <= n)
    return e;
  if (e.length === 3 && e[0] * e[1] <= n && e[2] <= n)
    return [e[0] * e[1], e[2]];
  if (e.length === 3 && e[0] <= n && e[1] * e[2] <= n)
    return [e[0], e[1] * e[2]];
  if (e.length === 4 && e[0] * e[1] * e[2] <= n && e[3] <= n)
    return [e[0] * e[1] * e[2], e[3]];
  if (e.length === 4 && e[0] <= n && e[1] * e[2] * e[3] <= n)
    return [e[0], e[1] * e[2] * e[3]];
  if (t) {
    let r = ya(e), a = 2, i = 2;
    return e.length && ([a, i] = va(e)), s = r * (a / 2) * (i / 2), w.sizeToSquarishShape(s).map((o) => o * 2);
  }
  return w.sizeToSquarishShape(s);
}
function ed(e) {
  return e % 2 === 0;
}
function rl(e, t) {
  if (e = e.slice(-2), t = t.slice(-2), w.arraysEqual(e, t) || !e.length || !t.length || e[0] === 0 || e[1] === 0 || t[0] === 0 || t[1] === 0)
    return true;
  if (e.length !== t.length) {
    let n = e.slice(-1)[0], s = t.slice(-1)[0];
    if (n === s || ed(n) && ed(s) && (e[0] === 1 || t[0] === 1))
      return true;
  }
  return e[1] === t[1] && ed(e[0]) && ed(t[0]);
}
var cd;
var dd;
function m1(e) {
  if (cd == null) {
    let t = xs(e);
    cd = t.getParameter(t.MAX_TEXTURE_SIZE);
  }
  return cd;
}
function rK() {
  cd = null;
}
function aK() {
  dd = null;
}
function g1(e) {
  if (dd == null) {
    let t = xs(e);
    dd = t.getParameter(t.MAX_TEXTURE_IMAGE_UNITS);
  }
  return Math.min(16, dd);
}
function b1(e) {
  if (e === 0)
    return 0;
  let t, n = xs(e);
  return Ln(n, "EXT_disjoint_timer_query_webgl2") && e === 2 ? t = 2 : Ln(n, "EXT_disjoint_timer_query") ? t = 1 : t = 0, t;
}
function Ln(e, t) {
  return e.getExtension(t) != null;
}
function Xm(e) {
  try {
    if (xs(e) != null)
      return true;
  } catch (t) {
    return console.log("Error when getting WebGL context: ", t), false;
  }
  return false;
}
function y1(e) {
  if (e === 0)
    return false;
  let t = xs(e);
  if (e === 1) {
    if (!Ln(t, "OES_texture_float"))
      return false;
  } else if (!Ln(t, "EXT_color_buffer_float"))
    return false;
  return Ym(t);
}
function v1(e) {
  if (e === 0)
    return false;
  let t = xs(e);
  if (e === 1) {
    if (!Ln(t, "OES_texture_float") || !Ln(t, "WEBGL_color_buffer_float"))
      return false;
  } else {
    if (Ln(t, "EXT_color_buffer_float"))
      return Ym(t);
    let s = "EXT_color_buffer_half_float";
    if (Ln(t, s)) {
      let r = t.getExtension(s);
      return iK(t, r);
    }
    return false;
  }
  return Ym(t);
}
function Ym(e) {
  let t = yv(e), n = e.createTexture();
  e.bindTexture(e.TEXTURE_2D, n);
  let s = 1, r = 1;
  e.texImage2D(e.TEXTURE_2D, 0, t.internalFormatFloat, s, r, 0, t.textureFormatFloat, t.textureTypeFloat, null);
  let a = e.createFramebuffer();
  e.bindFramebuffer(e.FRAMEBUFFER, a), e.framebufferTexture2D(e.FRAMEBUFFER, e.COLOR_ATTACHMENT0, e.TEXTURE_2D, n, 0);
  let i = e.checkFramebufferStatus(e.FRAMEBUFFER) === e.FRAMEBUFFER_COMPLETE;
  return e.bindTexture(e.TEXTURE_2D, null), e.bindFramebuffer(e.FRAMEBUFFER, null), e.deleteTexture(n), e.deleteFramebuffer(a), i;
}
function iK(e, t) {
  let n = yv(e, t), s = e.createTexture();
  e.bindTexture(e.TEXTURE_2D, s);
  let r = 1, a = 1;
  e.texImage2D(e.TEXTURE_2D, 0, n.internalFormatHalfFloat, r, a, 0, n.textureFormatFloat, n.textureTypeHalfFloat, null);
  let i = e.createFramebuffer();
  e.bindFramebuffer(e.FRAMEBUFFER, i), e.framebufferTexture2D(e.FRAMEBUFFER, e.COLOR_ATTACHMENT0, e.TEXTURE_2D, s, 0);
  let o = e.checkFramebufferStatus(e.FRAMEBUFFER) === e.FRAMEBUFFER_COMPLETE;
  return e.bindTexture(e.TEXTURE_2D, null), e.bindFramebuffer(e.FRAMEBUFFER, null), e.deleteTexture(s), e.deleteFramebuffer(i), o;
}
function x1(e) {
  return e !== 2 ? false : xs(e).fenceSync != null;
}
function iu(e, t) {
  Array.isArray(e) || (e = [e]), e.forEach((n) => {
    n != null && w.assert(n.dtype !== "complex64", () => `${t} does not support complex64 tensors in the WebGL backend.`);
  });
}
var Ne = K();
Ne.registerFlag("HAS_WEBGL", () => Ne.getNumber("WEBGL_VERSION") > 0);
Ne.registerFlag("WEBGL_VERSION", () => Xm(2) ? 2 : Xm(1) ? 1 : 0);
Ne.registerFlag("WEBGL_CHECK_NUMERICAL_PROBLEMS", () => false);
Ne.registerFlag("WEBGL_BUFFER_SUPPORTED", () => Ne.get("WEBGL_VERSION") === 2);
Ne.registerFlag("WEBGL_CPU_FORWARD", () => true);
Ne.registerFlag("WEBGL_FORCE_F16_TEXTURES", () => false);
Ne.registerFlag("WEBGL_PACK", () => Ne.getBool("HAS_WEBGL"));
Ne.registerFlag("WEBGL_PACK_NORMALIZATION", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_PACK_CLIP", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_PACK_DEPTHWISECONV", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_PACK_BINARY_OPERATIONS", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_PACK_UNARY_OPERATIONS", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_PACK_ARRAY_OPERATIONS", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_PACK_IMAGE_OPERATIONS", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_PACK_REDUCE", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_LAZILY_UNPACK", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_CONV_IM2COL", () => Ne.getBool("WEBGL_PACK"));
Ne.registerFlag("WEBGL_MAX_TEXTURE_SIZE", () => m1(Ne.getNumber("WEBGL_VERSION")));
Ne.registerFlag("WEBGL_MAX_TEXTURES_IN_SHADER", () => g1(Ne.getNumber("WEBGL_VERSION")));
Ne.registerFlag("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION", () => {
  let e = Ne.getNumber("WEBGL_VERSION");
  return e === 0 ? 0 : b1(e);
});
Ne.registerFlag("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE", () => Ne.getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION") > 0 && !yp.isMobile());
Ne.registerFlag("WEBGL_RENDER_FLOAT32_CAPABLE", () => y1(Ne.getNumber("WEBGL_VERSION")));
Ne.registerFlag("WEBGL_RENDER_FLOAT32_ENABLED", () => Ne.getBool("WEBGL_FORCE_F16_TEXTURES") ? false : Ne.getBool("WEBGL_RENDER_FLOAT32_CAPABLE"));
Ne.registerFlag("WEBGL_DOWNLOAD_FLOAT_ENABLED", () => v1(Ne.getNumber("WEBGL_VERSION")));
Ne.registerFlag("WEBGL_FENCE_API_ENABLED", () => x1(Ne.getNumber("WEBGL_VERSION")));
Ne.registerFlag("WEBGL_SIZE_UPLOAD_UNIFORM", () => Ne.getBool("WEBGL_RENDER_FLOAT32_ENABLED") ? 4 : 0);
Ne.registerFlag("WEBGL_DELETE_TEXTURE_THRESHOLD", () => -1, (e) => {
  if (e < 0 && e !== -1)
    throw new Error(`WEBGL_DELETE_TEXTURE_THRESHOLD must be -1 (indicating never delete) or at least 0, but got ${e}.`);
});
Ne.registerFlag("WEBGL_FLUSH_THRESHOLD", () => yp.isMobile() ? 1 : -1, (e) => {
  if (e < 0 && e !== -1)
    throw new Error(`WEBGL_FLUSH_THRESHOLD must be -1 (indicating never manual flush) or at least 0, but got ${e}.`);
});
Ne.registerFlag("CPU_HANDOFF_SIZE_THRESHOLD", () => 128);
Ne.registerFlag("WEBGL_USE_SHAPES_UNIFORMS", () => false);
Ne.registerFlag("TOPK_LAST_DIM_CPU_HANDOFF_SIZE_THRESHOLD", () => 1e5);
Ne.registerFlag("TOPK_K_CPU_HANDOFF_THRESHOLD", () => 128);
function fn() {
  let e, t, n, s, r, a, i, o, u, l;
  return K().getNumber("WEBGL_VERSION") === 2 ? (e = "#version 300 es", t = "in", n = "out", s = "in", r = "texture", a = "outputColor", i = "out vec4 outputColor;", o = `
      bool isnan_custom(float val) {
        uint floatToUint = floatBitsToUint(val);
        return (floatToUint & 0x7fffffffu) > 0x7f800000u;
      }

      bvec4 isnan_custom(vec4 val) {
        return bvec4(isnan_custom(val.x),
          isnan_custom(val.y), isnan_custom(val.z), isnan_custom(val.w));
      }

      #define isnan(value) isnan_custom(value)
    `, u = "", l = `
      #define round(value) newRound(value)
      int newRound(float value) {
        return int(floor(value + 0.5));
      }

      ivec4 newRound(vec4 value) {
        return ivec4(floor(value + vec4(0.5)));
      }
    `) : (e = "", t = "attribute", n = "varying", s = "varying", r = "texture2D", a = "gl_FragColor", i = "", o = `
      #define isnan(value) isnan_custom(value)
      bool isnan_custom(float val) {
        return (val > 0. || val < 1. || val == 0.) ? false : true;
      }
      bvec4 isnan_custom(vec4 val) {
        return bvec4(isnan(val.x), isnan(val.y), isnan(val.z), isnan(val.w));
      }
    `, u = `
      uniform float INFINITY;

      bool isinf(float val) {
        return abs(val) == INFINITY;
      }
      bvec4 isinf(vec4 val) {
        return equal(abs(val), vec4(INFINITY));
      }
    `, l = `
      int round(float value) {
        return int(floor(value + 0.5));
      }

      ivec4 round(vec4 value) {
        return ivec4(floor(value + vec4(0.5)));
      }
    `), { version: e, attribute: t, varyingVs: n, varyingFs: s, texture2D: r, output: a, defineOutput: i, defineSpecialNaN: o, defineSpecialInf: u, defineRound: l };
}
function wi(e, t, n = "index") {
  let s = w.computeStrides(t);
  return s.map((r, a) => {
    let i = `int ${e[a]} = ${n} / ${r}`, o = a === s.length - 1 ? `int ${e[a + 1]} = ${n} - ${e[a]} * ${r}` : `index -= ${e[a]} * ${r}`;
    return `${i}; ${o};`;
  }).join("");
}
function Jp(e, t, n = "index") {
  let s = w.computeStrides(t);
  return s.map((r, a) => {
    let i = `int ${e[a]} = ${n} / outShapeStrides[${a}]`, o = a === s.length - 1 ? `int ${e[a + 1]} = ${n} - ${e[a]} * outShapeStrides[${a}]` : `index -= ${e[a]} * outShapeStrides[${a}]`;
    return `${i}; ${o};`;
  }).join("");
}
function oK(e, t) {
  let n = e.length, s = e.map((a) => `${t}[${a}]`), r = new Array(n - 1);
  r[n - 2] = s[n - 1];
  for (let a = n - 3; a >= 0; --a)
    r[a] = `(${r[a + 1]} * ${s[a + 1]})`;
  return r;
}
function uK(e, t, n = "index") {
  let s = e.map((a, i) => i), r = oK(s, t);
  return r.map((a, i) => {
    let o = `int ${e[i]} = ${n} / ${r[i]}`, u = i === r.length - 1 ? `int ${e[i + 1]} = ${n} - ${e[i]} * ${r[i]}` : `index -= ${e[i]} * ${r[i]}`;
    return `${o}; ${u};`;
  }).join("");
}
function xv(e) {
  let t = w.computeStrides(e).map((n) => n.toString());
  return `
  int getFlatIndex(ivec3 coords) {
    return coords.x * ${t[0]} + coords.y * ${t[1]} + coords.z;
  }
`;
}
function wv() {
  return `
  int getFlatIndex(ivec3 coords) {
    return coords.x * outShapeStrides[0] + coords.y * outShapeStrides[1] + coords.z;
  }
`;
}
var w1 = `
  const float FLOAT_MAX = 1.70141184e38;
  const float FLOAT_MIN = 1.17549435e-38;

  lowp vec4 encode_float(highp float v) {
    if (isnan(v)) {
      return vec4(255, 255, 255, 255);
    }

    highp float av = abs(v);

    if(av < FLOAT_MIN) {
      return vec4(0.0, 0.0, 0.0, 0.0);
    } else if(v > FLOAT_MAX) {
      return vec4(0.0, 0.0, 128.0, 127.0) / 255.0;
    } else if(v < -FLOAT_MAX) {
      return vec4(0.0, 0.0,  128.0, 255.0) / 255.0;
    }

    highp vec4 c = vec4(0,0,0,0);

    highp float e = floor(log2(av));
    highp float m = exp2(fract(log2(av))) - 1.0;

    c[2] = floor(128.0 * m);
    m -= c[2] / 128.0;
    c[1] = floor(32768.0 * m);
    m -= c[1] / 32768.0;
    c[0] = floor(8388608.0 * m);

    highp float ebias = e + 127.0;
    c[3] = floor(ebias / 2.0);
    ebias -= c[3] * 2.0;
    c[2] += floor(ebias) * 128.0;

    c[3] += 128.0 * step(0.0, -v);

    return c / 255.0;
  }
`;
var { getBroadcastDims: k1 } = C;
function lK(e, t, n) {
  let s = [];
  if (e.forEach((h) => {
    let f = w.sizeFromShape(h.shapeInfo.logicalShape);
    if (h.shapeInfo.isUniform ? s.push(`uniform float ${h.name}${f > 1 ? `[${f}]` : ""};`) : (s.push(`uniform sampler2D ${h.name};`), s.push(`uniform int offset${h.name};`)), n.enableShapeUniforms) {
      let { uniformShape: m } = kv(n.packedInputs, h.shapeInfo.logicalShape, h.shapeInfo.texShape);
      switch (m.length) {
        case 1:
          s.push(`uniform int ${h.name}Shape;`);
          break;
        case 2:
          s.push(`uniform ivec2 ${h.name}Shape;`);
          break;
        case 3:
          s.push(`uniform ivec3 ${h.name}Shape;`);
          break;
        case 4:
          s.push(`uniform ivec4 ${h.name}Shape;`);
          break;
        default:
          break;
      }
      s.push(`uniform ivec2 ${h.name}TexShape;`);
    }
  }), n.enableShapeUniforms) {
    switch (t.logicalShape.length) {
      case 1:
        s.push("uniform int outShape;");
        break;
      case 2:
        s.push("uniform ivec2 outShape;"), s.push("uniform int outShapeStrides;");
        break;
      case 3:
        s.push("uniform ivec3 outShape;"), s.push("uniform ivec2 outShapeStrides;");
        break;
      case 4:
        s.push("uniform ivec4 outShape;"), s.push("uniform ivec3 outShapeStrides;");
        break;
      default:
        break;
    }
    s.push("uniform ivec2 outTexShape;");
  }
  n.customUniforms && n.customUniforms.forEach((h) => {
    s.push(`uniform ${h.type} ${h.name}${h.arrayIndex ? `[${h.arrayIndex}]` : ""};`);
  });
  let r = s.join(`
`), a = e.map((h) => cK(h, t, n.packedInputs, n.enableShapeUniforms)).join(`
`), i = t.texShape, o = fn(), u = hK(o), l, c, p = gK(o);
  return t.isPacked ? (l = dK(t.logicalShape, i, n.enableShapeUniforms), c = mK(o)) : (l = pK(t.logicalShape, i, n.enableShapeUniforms), c = fK(o)), n.packedInputs && (p += xK), [p, u, c, r, l, a, n.userCode].join(`
`);
}
function ou(e, t = false) {
  let n = e.shapeInfo.logicalShape;
  switch (n.length) {
    case 0:
      return RK(e, t);
    case 1:
      return FK(e, t);
    case 2:
      return PK(e, t);
    case 3:
      return MK(e, t);
    case 4:
      return BK(e, t);
    case 5:
      return VK(e);
    case 6:
      return WK(e);
    default:
      throw new Error(`${n.length}-D input sampling is not yet supported`);
  }
}
function S1(e, t) {
  switch (e.shapeInfo.logicalShape.length) {
    case 0:
      return EK(e);
    case 1:
      return DK(e, t);
    case 2:
      return OK(e, t);
    case 3:
      return zK(e, t);
    default:
      return LK(e, t);
  }
}
function cK(e, t, n = false, s) {
  let r = "";
  n ? r += S1(e, s) : r += ou(e, s);
  let a = e.shapeInfo.logicalShape, i = t.logicalShape;
  return a.length <= i.length && (n ? r += UK(e, t) : r += GK(e, t)), r;
}
function dK(e, t, n) {
  switch (e.length) {
    case 0:
      return I1();
    case 1:
      return wK(e, t, n);
    case 2:
      return _K(e, t, n);
    case 3:
      return SK(e, t, n);
    default:
      return CK(e, t, n);
  }
}
function pK(e, t, n) {
  switch (e.length) {
    case 0:
      return I1();
    case 1:
      return kK(e, t, n);
    case 2:
      return AK(e, t, n);
    case 3:
      return IK(e, t, n);
    case 4:
      return NK(e, t, n);
    case 5:
      return TK(e, t);
    case 6:
      return $K(e, t);
    default:
      throw new Error(`${e.length}-D output sampling is not yet supported`);
  }
}
function hK(e) {
  return `
    float sampleTexture(sampler2D textureSampler, vec2 uv) {
      return ${e.texture2D}(textureSampler, uv).r;
    }
  `;
}
function fK(e) {
  return `
    void setOutput(float val) {
      ${e.output} = vec4(val, 0, 0, 0);
    }
  `;
}
function mK(e) {
  return `
    void setOutput(vec4 val) {
      ${e.output} = val;
    }
  `;
}
function gK(e) {
  return `${e.version}
    precision highp float;
    precision highp int;
    precision highp sampler2D;
    ${e.varyingFs} vec2 resultUV;
    ${e.defineOutput}
    const vec2 halfCR = vec2(0.5, 0.5);

    struct ivec5
    {
      int x;
      int y;
      int z;
      int w;
      int u;
    };

    struct ivec6
    {
      int x;
      int y;
      int z;
      int w;
      int u;
      int v;
    };

    uniform float NAN;
    ${e.defineSpecialNaN}
    ${e.defineSpecialInf}
    ${e.defineRound}

    int imod(int x, int y) {
      return x - y * (x / y);
    }

    int idiv(int a, int b, float sign) {
      int res = a / b;
      int mod = imod(a, b);
      if (sign < 0. && mod != 0) {
        res -= 1;
      }
      return res;
    }

    //Based on the work of Dave Hoskins
    //https://www.shadertoy.com/view/4djSRW
    #define HASHSCALE1 443.8975
    float random(float seed){
      vec2 p = resultUV * seed;
      vec3 p3  = fract(vec3(p.xyx) * HASHSCALE1);
      p3 += dot(p3, p3.yzx + 19.19);
      return fract((p3.x + p3.y) * p3.z);
    }

    ${bK}
    ${yK}
    ${vK}
  `;
}
var bK = `
vec2 uvFromFlat(int texNumR, int texNumC, int index) {
  int texR = index / texNumC;
  int texC = index - texR * texNumC;
  return (vec2(texC, texR) + halfCR) / vec2(texNumC, texNumR);
}
vec2 packedUVfrom1D(int texNumR, int texNumC, int index) {
  int texelIndex = index / 2;
  int texR = texelIndex / texNumC;
  int texC = texelIndex - texR * texNumC;
  return (vec2(texC, texR) + halfCR) / vec2(texNumC, texNumR);
}
`;
var yK = `
vec2 packedUVfrom2D(int texelsInLogicalRow, int texNumR,
  int texNumC, int row, int col) {
  int texelIndex = (row / 2) * texelsInLogicalRow + (col / 2);
  int texR = texelIndex / texNumC;
  int texC = texelIndex - texR * texNumC;
  return (vec2(texC, texR) + halfCR) / vec2(texNumC, texNumR);
}
`;
var vK = `
vec2 packedUVfrom3D(int texNumR, int texNumC,
    int texelsInBatch, int texelsInLogicalRow, int b,
    int row, int col) {
  int index = b * texelsInBatch + (row / 2) * texelsInLogicalRow + (col / 2);
  int texR = index / texNumC;
  int texC = index - texR * texNumC;
  return (vec2(texC, texR) + halfCR) / vec2(texNumC, texNumR);
}
`;
var xK = `
  float getChannel(vec4 frag, vec2 innerDims) {
    vec2 modCoord = mod(innerDims, 2.);
    return modCoord.x == 0. ?
      (modCoord.y == 0. ? frag.r : frag.g) :
      (modCoord.y == 0. ? frag.b : frag.a);
  }
  float getChannel(vec4 frag, int dim) {
    float modCoord = mod(float(dim), 2.);
    return modCoord == 0. ? frag.r : frag.g;
  }
`;
function I1() {
  return `
    int getOutputCoords() {
      return 0;
    }
  `;
}
function wK(e, t, n) {
  let s = [Math.ceil(t[0] / 2), Math.ceil(t[1] / 2)];
  return s[0] === 1 ? n ? `
      int getOutputCoords() {
        return 2 * int(resultUV.x * ceil(float(outTexShape[1]) / 2.0));
      }
    ` : `
      int getOutputCoords() {
        return 2 * int(resultUV.x * ${s[1]}.0);
      }
    ` : s[1] === 1 ? n ? `
      int getOutputCoords() {
        return 2 * int(resultUV.y * ceil(float(outTexShape[0]) / 2.0));
      }
    ` : `
      int getOutputCoords() {
        return 2 * int(resultUV.y * ${s[0]}.0);
      }
    ` : n ? `
    int getOutputCoords() {
      ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(packedTexShape[0], packedTexShape[1]));
      return 2 * (resTexRC.x * packedTexShape[1] + resTexRC.y);
    }
  ` : `
    int getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(${s[0]}, ${s[1]}));
      return 2 * (resTexRC.x * ${s[1]} + resTexRC.y);
    }
  `;
}
function kK(e, t, n) {
  return t[0] === 1 ? n ? `
      int getOutputCoords() {
        return int(resultUV.x * float(outTexShape[1]));
      }
    ` : `
      int getOutputCoords() {
        return int(resultUV.x * ${t[1]}.0);
      }
    ` : t[1] === 1 ? n ? `
      int getOutputCoords() {
        return int(resultUV.y * float(outTexShape[0]));
      }
    ` : `
      int getOutputCoords() {
        return int(resultUV.y * ${t[0]}.0);
      }
    ` : n ? `
    int getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(outTexShape[0], outTexShape[1]));
      return resTexRC.x * outTexShape[1] + resTexRC.y;
    }
  ` : `
    int getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(${t[0]}, ${t[1]}));
      return resTexRC.x * ${t[1]} + resTexRC.y;
    }
  `;
}
function SK(e, t, n) {
  if (n)
    return `
    ivec3 getOutputCoords() {
      ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
      int texelsInLogicalRow = int(ceil(float(outShape[2]) / 2.0));
      int texelsInBatch = texelsInLogicalRow * int(ceil(float(outShape[1]) / 2.0));
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(packedTexShape[0], packedTexShape[1]));
      int index = resTexRC.x * packedTexShape[1] + resTexRC.y;

      int b = index / texelsInBatch;
      index -= b * texelsInBatch;

      int r = 2 * (index / texelsInLogicalRow);
      int c = imod(index, texelsInLogicalRow) * 2;

      return ivec3(b, r, c);
    }
  `;
  let s = [Math.ceil(t[0] / 2), Math.ceil(t[1] / 2)], r = Math.ceil(e[2] / 2), a = r * Math.ceil(e[1] / 2);
  return `
    ivec3 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(${s[0]}, ${s[1]}));
      int index = resTexRC.x * ${s[1]} + resTexRC.y;

      int b = index / ${a};
      index -= b * ${a};

      int r = 2 * (index / ${r});
      int c = imod(index, ${r}) * 2;

      return ivec3(b, r, c);
    }
  `;
}
function IK(e, t, n) {
  if (n)
    return `
  ivec3 getOutputCoords() {
    ivec2 resTexRC = ivec2(resultUV.yx *
                           vec2(outTexShape[0], outTexShape[1]));
    int index = resTexRC.x * outTexShape[1] + resTexRC.y;
    ${Jp(["r", "c", "d"], e)}
    return ivec3(r, c, d);
  }
`;
  let s = wi(["r", "c", "d"], e);
  return `
    ivec3 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(${t[0]}, ${t[1]}));
      int index = resTexRC.x * ${t[1]} + resTexRC.y;
      ${s}
      return ivec3(r, c, d);
    }
  `;
}
function CK(e, t, n) {
  if (n)
    return `
    ivec4 getOutputCoords() {
      ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(packedTexShape[0], packedTexShape[1]));
      int index = resTexRC.x * packedTexShape[1] + resTexRC.y;

      int texelsInLogicalRow = int(ceil(float(outShape[3]) / 2.0));
      int texelsInBatch = texelsInLogicalRow * int(ceil(float(outShape[2]) / 2.0));
      int texelsInBatchN = texelsInBatch * outShape[1];

      int b2 = index / texelsInBatchN;
      index -= b2 * texelsInBatchN;

      int b = index / texelsInBatch;
      index -= b * texelsInBatch;

      int r = 2 * (index / texelsInLogicalRow);
      int c = imod(index, texelsInLogicalRow) * 2;

      return ivec4(b2, b, r, c);
    }
  `;
  let s = [Math.ceil(t[0] / 2), Math.ceil(t[1] / 2)], r = Math.ceil(e[e.length - 1] / 2), a = r * Math.ceil(e[e.length - 2] / 2), i = a, o = "", u = "b, r, c";
  for (let l = 2; l < e.length - 1; l++)
    i *= e[e.length - l - 1], o = `
      int b${l} = index / ${i};
      index -= b${l} * ${i};
    ` + o, u = `b${l}, ` + u;
  return `
    ivec${e.length} getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(${s[0]}, ${s[1]}));
      int index = resTexRC.x * ${s[1]} + resTexRC.y;

      ${o}

      int b = index / ${a};
      index -= b * ${a};

      int r = 2 * (index / ${r});
      int c = imod(index, ${r}) * 2;

      return ivec${e.length}(${u});
    }
  `;
}
function NK(e, t, n) {
  if (n)
    return `
    ivec4 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
        vec2(outTexShape[0], outTexShape[1]));
      int index = resTexRC.x * outTexShape[1] + resTexRC.y;
      ${Jp(["r", "c", "d", "d2"], e)}
      return ivec4(r, c, d, d2);
    }
  `;
  let s = wi(["r", "c", "d", "d2"], e);
  return `
    ivec4 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
        vec2(${t[0]}, ${t[1]}));
      int index = resTexRC.x * ${t[1]} + resTexRC.y;
      ${s}
      return ivec4(r, c, d, d2);
    }
  `;
}
function TK(e, t) {
  let n = wi(["r", "c", "d", "d2", "d3"], e);
  return `
    ivec5 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx * vec2(${t[0]},
                             ${t[1]}));

      int index = resTexRC.x * ${t[1]} + resTexRC.y;

      ${n}

      ivec5 outShape = ivec5(r, c, d, d2, d3);
      return outShape;
    }
  `;
}
function $K(e, t) {
  let n = wi(["r", "c", "d", "d2", "d3", "d4"], e);
  return `
    ivec6 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
        vec2(${t[0]}, ${t[1]}));
      int index = resTexRC.x * ${t[1]} + resTexRC.y;

      ${n}

      ivec6 result = ivec6(r, c, d, d2, d3, d4);
      return result;
    }
  `;
}
function _K(e, t, n) {
  let s = [Math.ceil(t[0] / 2), Math.ceil(t[1] / 2)];
  if (w.arraysEqual(e, t))
    return n ? `
      ivec2 getOutputCoords() {
        ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
        return 2 * ivec2(resultUV.yx * vec2(packedTexShape[0], packedTexShape[1]));
      }
    ` : `
      ivec2 getOutputCoords() {
        return 2 * ivec2(resultUV.yx * vec2(${s[0]}, ${s[1]}));
      }
    `;
  let r = Math.ceil(e[1] / 2);
  return n ? `
    ivec2 getOutputCoords() {
      ivec2 packedTexShape = ivec2(ceil(float(outTexShape[0]) / 2.0), ceil(float(outTexShape[1]) / 2.0));
      int texelsInLogicalRow = int(ceil(float(outShape[1]) / 2.0));
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(packedTexShape[0], packedTexShape[1]));

      int index = resTexRC.x * packedTexShape[1] + resTexRC.y;
      int r = 2 * (index / texelsInLogicalRow);
      int c = imod(index, texelsInLogicalRow) * 2;

      return ivec2(r, c);
    }
  ` : `
    ivec2 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(${s[0]}, ${s[1]}));

      int index = resTexRC.x * ${s[1]} + resTexRC.y;
      int r = 2 * (index / ${r});
      int c = imod(index, ${r}) * 2;

      return ivec2(r, c);
    }
  `;
}
function AK(e, t, n) {
  return w.arraysEqual(e, t) ? n ? `
      ivec2 getOutputCoords() {
        return ivec2(resultUV.yx * vec2(outTexShape[0], outTexShape[1]));
      }
    ` : `
      ivec2 getOutputCoords() {
        return ivec2(resultUV.yx * vec2(${t[0]}, ${t[1]}));
      }
    ` : e[1] === 1 ? n ? `
      ivec2 getOutputCoords() {
        ivec2 resTexRC = ivec2(resultUV.yx *
                               vec2(outTexShape[0], outTexShape[1]));
        int index = resTexRC.x * outTexShape[1] + resTexRC.y;
        return ivec2(index, 0);
      }
    ` : `
      ivec2 getOutputCoords() {
        ivec2 resTexRC = ivec2(resultUV.yx *
                               vec2(${t[0]}, ${t[1]}));
        int index = resTexRC.x * ${t[1]} + resTexRC.y;
        return ivec2(index, 0);
      }
    ` : e[0] === 1 ? n ? `
      ivec2 getOutputCoords() {
        ivec2 resTexRC = ivec2(resultUV.yx *
                               vec2(outTexShape[0], outTexShape[1]));
        int index = resTexRC.x * outTexShape[1] + resTexRC.y;
        return ivec2(0, index);
      }
    ` : `
      ivec2 getOutputCoords() {
        ivec2 resTexRC = ivec2(resultUV.yx *
                               vec2(${t[0]}, ${t[1]}));
        int index = resTexRC.x * ${t[1]} + resTexRC.y;
        return ivec2(0, index);
      }
    ` : n ? `
    ivec2 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(outTexShape[0], outTexShape[1]));
      int index = resTexRC.x * outTexShape[1] + resTexRC.y;
      int r = index / outShape[1];
      int c = index - r * outShape[1];
      return ivec2(r, c);
    }
  ` : `
    ivec2 getOutputCoords() {
      ivec2 resTexRC = ivec2(resultUV.yx *
                             vec2(${t[0]}, ${t[1]}));
      int index = resTexRC.x * ${t[1]} + resTexRC.y;
      int r = index / ${e[1]};
      int c = index - r * ${e[1]};
      return ivec2(r, c);
    }
  `;
}
function ki(e) {
  return `offset${e}`;
}
function EK(e) {
  let t = e.name, n = "get" + t.charAt(0).toUpperCase() + t.slice(1), s = fn();
  return `
    vec4 ${n}() {
      return ${s.texture2D}(${t}, halfCR);
    }
  `;
}
function RK(e, t) {
  let n = e.name, s = "get" + n.charAt(0).toUpperCase() + n.slice(1);
  if (e.shapeInfo.isUniform)
    return `float ${s}() {return ${n};}`;
  let [r, a] = e.shapeInfo.texShape;
  if (r === 1 && a === 1)
    return `
      float ${s}() {
        return sampleTexture(${n}, halfCR);
      }
    `;
  let i = ki(n);
  if (t)
    return `
    float ${s}() {
      vec2 uv = uvFromFlat(${n}TexShape[0], ${n}TexShape[1], ${i});
      return sampleTexture(${n}, uv);
    }
  `;
  let [o, u] = e.shapeInfo.texShape;
  return `
    float ${s}() {
      vec2 uv = uvFromFlat(${o}, ${u}, ${i});
      return sampleTexture(${n}, uv);
    }
  `;
}
function DK(e, t) {
  let n = e.name, s = "get" + n.charAt(0).toUpperCase() + n.slice(1), r = e.shapeInfo.texShape, a = fn();
  if (t)
    return `
    vec4 ${s}(int index) {
      ivec2 packedTexShape = ivec2(ceil(float(${n}TexShape[0]) / 2.0), ceil(float(${n}TexShape[1]) / 2.0));
      vec2 uv = packedUVfrom1D(
        packedTexShape[0], packedTexShape[1], index);
      return ${a.texture2D}(${n}, uv);
    }
  `;
  let i = [Math.ceil(r[0] / 2), Math.ceil(r[1] / 2)];
  return `
    vec4 ${s}(int index) {
      vec2 uv = packedUVfrom1D(
        ${i[0]}, ${i[1]}, index);
      return ${a.texture2D}(${n}, uv);
    }
  `;
}
function FK(e, t) {
  let n = e.name, s = "get" + n.charAt(0).toUpperCase() + n.slice(1);
  if (e.shapeInfo.isUniform)
    return `
      float ${s}(int index) {
        ${uu(e)}
      }
    `;
  let r = e.shapeInfo.texShape, a = r[0], i = r[1];
  if (i === 1 && a === 1)
    return `
      float ${s}(int index) {
        return sampleTexture(${n}, halfCR);
      }
    `;
  let o = ki(n);
  return i === 1 ? t ? `
      float ${s}(int index) {
        vec2 uv = vec2(0.5, (float(index + ${o}) + 0.5) / float(${n}TexShape[0]));
        return sampleTexture(${n}, uv);
      }
    ` : `
      float ${s}(int index) {
        vec2 uv = vec2(0.5, (float(index + ${o}) + 0.5) / ${a}.0);
        return sampleTexture(${n}, uv);
      }
    ` : a === 1 ? t ? `
      float ${s}(int index) {
        vec2 uv = vec2((float(index + ${o}) + 0.5) / float(${n}TexShape[1]), 0.5);
        return sampleTexture(${n}, uv);
      }
    ` : `
      float ${s}(int index) {
        vec2 uv = vec2((float(index + ${o}) + 0.5) / ${i}.0, 0.5);
        return sampleTexture(${n}, uv);
      }
    ` : t ? `
    float ${s}(int index) {
      vec2 uv = uvFromFlat(${n}TexShape[0], ${n}TexShape[1], index + ${o});
      return sampleTexture(${n}, uv);
    }
  ` : `
    float ${s}(int index) {
      vec2 uv = uvFromFlat(${a}, ${i}, index + ${o});
      return sampleTexture(${n}, uv);
    }
  `;
}
function OK(e, t) {
  let n = e.shapeInfo.logicalShape, s = e.name, r = "get" + s.charAt(0).toUpperCase() + s.slice(1), a = e.shapeInfo.texShape, i = a[0], o = a[1], u = fn();
  if (a != null && w.arraysEqual(n, a))
    return t ? `
      vec4 ${r}(int row, int col) {
        vec2 uv = (vec2(col, row) + halfCR) / vec2(${s}TexShape[1], ${s}TexShape[0]);

        return ${u.texture2D}(${s}, uv);
      }
    ` : `
      vec4 ${r}(int row, int col) {
        vec2 uv = (vec2(col, row) + halfCR) / vec2(${o}.0, ${i}.0);

        return ${u.texture2D}(${s}, uv);
      }
    `;
  if (t)
    return `
    vec4 ${r}(int row, int col) {
      ivec2 packedTexShape = ivec2(ceil(float(${s}TexShape[0]) / 2.0), ceil(float(${s}TexShape[1]) / 2.0));
      int valuesPerRow = int(ceil(float(${s}Shape[1]) / 2.0));
      vec2 uv = packedUVfrom2D(valuesPerRow, packedTexShape[0], packedTexShape[1], row, col);
      return ${u.texture2D}(${s}, uv);
    }
  `;
  let l = [Math.ceil(a[0] / 2), Math.ceil(a[1] / 2)], c = Math.ceil(n[1] / 2);
  return `
    vec4 ${r}(int row, int col) {
      vec2 uv = packedUVfrom2D(${c}, ${l[0]}, ${l[1]}, row, col);
      return ${u.texture2D}(${s}, uv);
    }
  `;
}
function PK(e, t) {
  let n = e.shapeInfo.logicalShape, s = e.name, r = "get" + s.charAt(0).toUpperCase() + s.slice(1), a = e.shapeInfo.texShape;
  if (a != null && w.arraysEqual(n, a)) {
    if (t)
      return `
      float ${r}(int row, int col) {
        vec2 uv = (vec2(col, row) + halfCR) / vec2(${s}TexShape[1], ${s}TexShape[0]);
        return sampleTexture(${s}, uv);
      }
    `;
    let d = a[0], h = a[1];
    return `
    float ${r}(int row, int col) {
      vec2 uv = (vec2(col, row) + halfCR) / vec2(${h}.0, ${d}.0);
      return sampleTexture(${s}, uv);
    }
  `;
  }
  let { newShape: i, keptDims: o } = w.squeezeShape(n), u = i;
  if (u.length < n.length) {
    let d = lu(e, u), h = ["row", "col"];
    return `
      ${ou(d, t)}
      float ${r}(int row, int col) {
        return ${r}(${cu(h, o)});
      }
    `;
  }
  if (e.shapeInfo.isUniform)
    return `
      float ${r}(int row, int col) {
        int index = round(dot(vec2(row, col), vec2(${n[1]}, 1)));
        ${uu(e)}
      }
    `;
  let l = a[0], c = a[1], p = ki(s);
  return c === 1 ? t ? `
      float ${r}(int row, int col) {
        float index = dot(vec3(row, col, ${p}), vec3(${s}Shape[1], 1, 1));
        vec2 uv = vec2(0.5, (index + 0.5) / float(${s}TexShape[0]));
        return sampleTexture(${s}, uv);
      }
    ` : `
    float ${r}(int row, int col) {
      float index = dot(vec3(row, col, ${p}), vec3(${n[1]}, 1, 1));
      vec2 uv = vec2(0.5, (index + 0.5) / ${l}.0);
      return sampleTexture(${s}, uv);
    }
  ` : l === 1 ? t ? `
      float ${r}(int row, int col) {
        float index = dot(vec3(row, col, ${p}), vec3(${s}Shape[1], 1, 1));
        vec2 uv = vec2((index + 0.5) / float(${s}TexShape[1]), 0.5);
        return sampleTexture(${s}, uv);
      }
    ` : `
    float ${r}(int row, int col) {
      float index = dot(vec3(row, col, ${p}), vec3(${n[1]}, 1, 1));
      vec2 uv = vec2((index + 0.5) / ${c}.0, 0.5);
      return sampleTexture(${s}, uv);
    }
  ` : t ? `
      float ${r}(int row, int col) {
        // Explicitly use integer operations as dot() only works on floats.
        int index = row * ${s}Shape[1] + col + ${p};
        vec2 uv = uvFromFlat(${s}TexShape[0], ${s}TexShape[1], index);
        return sampleTexture(${s}, uv);
      }
    ` : `
  float ${r}(int row, int col) {
    // Explicitly use integer operations as dot() only works on floats.
    int index = row * ${n[1]} + col + ${p};
    vec2 uv = uvFromFlat(${l}, ${c}, index);
    return sampleTexture(${s}, uv);
  }
`;
}
function zK(e, t) {
  let n = e.shapeInfo.logicalShape, s = e.name, r = "get" + s.charAt(0).toUpperCase() + s.slice(1), a = e.shapeInfo.texShape, i = [Math.ceil(a[0] / 2), Math.ceil(a[1] / 2)];
  if (n[0] === 1) {
    let d = n.slice(1), h = [1, 2], f = lu(e, d), m = ["b", "row", "col"];
    return `
        ${S1(f, t)}
        vec4 ${r}(int b, int row, int col) {
          return ${r}(${cu(m, h)});
        }
      `;
  }
  let o = fn();
  if (t)
    return `
    vec4 ${r}(int b, int row, int col) {
      ivec2 packedTexShape = ivec2(ceil(float(${s}TexShape[0]) / 2.0), ceil(float(${s}TexShape[1]) / 2.0));
      int valuesPerRow = int(ceil(float(${s}Shape[2]) / 2.0));
      int texelsInBatch = valuesPerRow * int(ceil(float(${s}Shape[1]) / 2.0));
      vec2 uv = packedUVfrom3D(
        packedTexShape[0], packedTexShape[1], texelsInBatch, valuesPerRow, b, row, col);
      return ${o.texture2D}(${s}, uv);
    }
  `;
  let u = i[0], l = i[1], c = Math.ceil(n[2] / 2), p = c * Math.ceil(n[1] / 2);
  return `
    vec4 ${r}(int b, int row, int col) {
      vec2 uv = packedUVfrom3D(
        ${u}, ${l}, ${p}, ${c}, b, row, col);
      return ${o.texture2D}(${s}, uv);
    }
  `;
}
function MK(e, t) {
  let n = e.shapeInfo.logicalShape, s = e.name, r = "get" + s.charAt(0).toUpperCase() + s.slice(1), a = n[1] * n[2], i = n[2], { newShape: o, keptDims: u } = w.squeezeShape(n), l = o;
  if (l.length < n.length) {
    let m = lu(e, l), g = ["row", "col", "depth"];
    return `
        ${ou(m, t)}
        float ${r}(int row, int col, int depth) {
          return ${r}(${cu(g, u)});
        }
      `;
  }
  if (e.shapeInfo.isUniform)
    return `
      float ${r}(int row, int col, int depth) {
        int index = round(dot(vec3(row, col, depth),
                          vec3(${a}, ${i}, 1)));
        ${uu(e)}
      }
    `;
  let c = e.shapeInfo.texShape, p = c[0], d = c[1], h = e.shapeInfo.flatOffset;
  if (d === a && h == null)
    return t ? `
      float ${r}(int row, int col, int depth) {
        int stride1 = ${s}Shape[2];
        float texR = float(row);
        float texC = dot(vec2(col, depth), vec2(stride1, 1));
        vec2 uv = (vec2(texC, texR) + halfCR) /
                   vec2(${s}TexShape[1], ${s}TexShape[0]);
        return sampleTexture(${s}, uv);
      }
    ` : `
        float ${r}(int row, int col, int depth) {
          float texR = float(row);
          float texC = dot(vec2(col, depth), vec2(${i}, 1));
          vec2 uv = (vec2(texC, texR) + halfCR) /
                     vec2(${d}.0, ${p}.0);
          return sampleTexture(${s}, uv);
        }
      `;
  if (d === i && h == null)
    return t ? `
      float ${r}(int row, int col, int depth) {
        float texR = dot(vec2(row, col), vec2(${s}Shape[1], 1));
        float texC = float(depth);
        vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${s}TexShape[1], ${s}TexShape[0]);
        return sampleTexture(${s}, uv);
      }
    ` : `
    float ${r}(int row, int col, int depth) {
      float texR = dot(vec2(row, col), vec2(${n[1]}, 1));
      float texC = float(depth);
      vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${d}.0, ${p}.0);
      return sampleTexture(${s}, uv);
    }
  `;
  let f = ki(s);
  return t ? `
    float ${r}(int row, int col, int depth) {
      // Explicitly use integer operations as dot() only works on floats.
      int stride0 = ${s}Shape[1] * ${s}Shape[2];
      int stride1 = ${s}Shape[2];
      int index = row * ${a} + col * ${i} + depth + ${f};
      vec2 uv = uvFromFlat(${s}TexShape[0], ${s}TexShape[1], index);
      return sampleTexture(${s}, uv);
    }
    ` : `
      float ${r}(int row, int col, int depth) {
        // Explicitly use integer operations as dot() only works on floats.
        int index = row * ${a} + col * ${i} + depth + ${f};
        vec2 uv = uvFromFlat(${p}, ${d}, index);
        return sampleTexture(${s}, uv);
      }
  `;
}
function LK(e, t) {
  let n = e.name, s = "get" + n.charAt(0).toUpperCase() + n.slice(1), r = fn();
  if (t)
    return `
    vec4 ${s}(int b2, int b, int row, int col) {
      int valuesPerRow = int(ceil(float(${n}Shape[3]) / 2.0));
      int texelsInBatch = valuesPerRow * int(ceil(float(${n}Shape[2]) / 2.0));
      int index = b * texelsInBatch + (row / 2) * valuesPerRow + (col / 2);
      texelsInBatch *= ${n}Shape[1];
      index = b2 * texelsInBatch + index;
      ivec2 packedTexShape = ivec2(ceil(float(${n}TexShape[0]) / 2.0), ceil(float(${n}TexShape[1]) / 2.0));
      int texR = index / packedTexShape[1];
      int texC = index - texR * packedTexShape[1];
      vec2 uv = (vec2(texC, texR) + halfCR) / vec2(packedTexShape[1], packedTexShape[0]); return ${r.texture2D}(${n}, uv);
    }
  `;
  let a = e.shapeInfo.logicalShape, i = a.length, o = e.shapeInfo.texShape, u = [Math.ceil(o[0] / 2), Math.ceil(o[1] / 2)], l = u[0], c = u[1], p = Math.ceil(a[i - 1] / 2), d = p * Math.ceil(a[i - 2] / 2), h = "int b, int row, int col", f = `b * ${d} + (row / 2) * ${p} + (col / 2)`;
  for (let m = 2; m < i - 1; m++)
    h = `int b${m}, ` + h, d *= a[i - m - 1], f = `b${m} * ${d} + ` + f;
  return `
    vec4 ${s}(${h}) {
      int index = ${f};
      int texR = index / ${c};
      int texC = index - texR * ${c};
      vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${c}, ${l});
      return ${r.texture2D}(${n}, uv);
    }
  `;
}
function BK(e, t) {
  let n = e.shapeInfo.logicalShape, s = e.name, r = "get" + s.charAt(0).toUpperCase() + s.slice(1), a = n[3], i = n[2] * a, o = n[1] * i, { newShape: u, keptDims: l } = w.squeezeShape(n);
  if (u.length < n.length) {
    let y = lu(e, u), v = ["row", "col", "depth", "depth2"];
    return `
      ${ou(y, t)}
      float ${r}(int row, int col, int depth, int depth2) {
        return ${r}(${cu(v, l)});
      }
    `;
  }
  if (e.shapeInfo.isUniform)
    return `
      float ${r}(int row, int col, int depth, int depth2) {
        int index = round(dot(vec4(row, col, depth, depth2),
                          vec4(${o}, ${i}, ${a}, 1)));
        ${uu(e)}
      }
    `;
  let c = e.shapeInfo.flatOffset, p = e.shapeInfo.texShape, d = p[0], h = p[1], f = `int stride2 = ${s}Shape[3];`, m = `int stride1 = ${s}Shape[2] * stride2;`, g = `int stride0 = ${s}Shape[1] * stride1;`;
  if (h === o && c == null)
    return t ? `
      float ${r}(int row, int col, int depth, int depth2) {
        ${f}
        ${m}
        float texR = float(row);
        float texC =
            dot(vec3(col, depth, depth2),
                vec3(stride1, stride2, 1));
        vec2 uv = (vec2(texC, texR) + halfCR) /
                   vec2(${s}TexShape[1], ${s}TexShape[0]);
        return sampleTexture(${s}, uv);
      }
    ` : `
      float ${r}(int row, int col, int depth, int depth2) {
        float texR = float(row);
        float texC =
            dot(vec3(col, depth, depth2),
                vec3(${i}, ${a}, 1));
        vec2 uv = (vec2(texC, texR) + halfCR) /
                   vec2(${h}.0, ${d}.0);
        return sampleTexture(${s}, uv);
      }
    `;
  if (h === a && c == null)
    return t ? `
      float ${r}(int row, int col, int depth, int depth2) {
        float texR = dot(vec3(row, col, depth),
                         vec3(${s}Shape[1] * ${s}Shape[2], ${s}Shape[2], 1));
        float texC = float(depth2);
        vec2 uv = (vec2(texC, texR) + halfCR) /
                  vec2(${s}TexShape[1], ${s}TexShape[0]);
        return sampleTexture(${s}, uv);
      }
    ` : `
      float ${r}(int row, int col, int depth, int depth2) {
        float texR = dot(vec3(row, col, depth),
                         vec3(${n[1] * n[2]}, ${n[2]}, 1));
        float texC = float(depth2);
        vec2 uv = (vec2(texC, texR) + halfCR) /
                  vec2(${h}.0, ${d}.0);
        return sampleTexture(${s}, uv);
      }
    `;
  let b = ki(s);
  return t ? `
    float ${r}(int row, int col, int depth, int depth2) {
      // Explicitly use integer operations as dot() only works on floats.
      ${f}
      ${m}
      ${g}
      int index = row * stride0 + col * stride1 +
          depth * stride2 + depth2;
      vec2 uv = uvFromFlat(${s}TexShape[0], ${s}TexShape[1], index + ${b});
      return sampleTexture(${s}, uv);
    }
  ` : `
    float ${r}(int row, int col, int depth, int depth2) {
      // Explicitly use integer operations as dot() only works on floats.
      int index = row * ${o} + col * ${i} +
          depth * ${a} + depth2;
      vec2 uv = uvFromFlat(${d}, ${h}, index + ${b});
      return sampleTexture(${s}, uv);
    }
  `;
}
function VK(e) {
  let t = e.shapeInfo.logicalShape, n = e.name, s = "get" + n.charAt(0).toUpperCase() + n.slice(1), r = t[4], a = t[3] * r, i = t[2] * a, o = t[1] * i, { newShape: u, keptDims: l } = w.squeezeShape(t);
  if (u.length < t.length) {
    let m = lu(e, u), g = ["row", "col", "depth", "depth2", "depth3"];
    return `
      ${ou(m)}
      float ${s}(int row, int col, int depth, int depth2, int depth3) {
        return ${s}(${cu(g, l)});
      }
    `;
  }
  if (e.shapeInfo.isUniform)
    return `
      float ${s}(int row, int col, int depth, int depth2, int depth3) {
        float index = dot(
          vec4(row, col, depth, depth2),
          vec4(${o}, ${i}, ${a}, ${r})) +
          depth3;
        ${uu(e)}
      }
    `;
  let c = e.shapeInfo.flatOffset, p = e.shapeInfo.texShape, d = p[0], h = p[1];
  if (h === o && c == null)
    return `
      float ${s}(int row, int col, int depth, int depth2, int depth3) {
        int texR = row;
        float texC = dot(vec4(col, depth, depth2, depth3),
                         vec4(${i}, ${a}, ${r}, 1));
        vec2 uv = (vec2(texC, texR) + halfCR) /
                   vec2(${h}.0, ${d}.0);
        return sampleTexture(${n}, uv);
      }
    `;
  if (h === r && c == null)
    return `
      float ${s}(int row, int col, int depth, int depth2, int depth3) {
        float texR = dot(
          vec4(row, col, depth, depth2),
          vec4(${t[1] * t[2] * t[3]},
               ${t[2] * t[3]}, ${t[3]}, 1));
        int texC = depth3;
        vec2 uv = (vec2(texC, texR) + halfCR) /
                  vec2(${h}.0, ${d}.0);
        return sampleTexture(${n}, uv);
      }
    `;
  let f = ki(n);
  return `
    float ${s}(int row, int col, int depth, int depth2, int depth3) {
      // Explicitly use integer operations as dot() only works on floats.
      int index = row * ${o} + col * ${i} + depth * ${a} +
          depth2 * ${r} + depth3 + ${f};
      vec2 uv = uvFromFlat(${d}, ${h}, index);
      return sampleTexture(${n}, uv);
    }
  `;
}
function WK(e) {
  let t = e.shapeInfo.logicalShape, n = e.name, s = "get" + n.charAt(0).toUpperCase() + n.slice(1), { newShape: r, keptDims: a } = w.squeezeShape(t);
  if (r.length < t.length) {
    let g = lu(e, r), b = ["row", "col", "depth", "depth2", "depth3", "depth4"];
    return `
      ${ou(g)}
      float ${s}(int row, int col, int depth,
                    int depth2, int depth3, int depth4) {
        return ${s}(${cu(b, a)});
      }
    `;
  }
  let i = t[5], o = t[4] * i, u = t[3] * o, l = t[2] * u, c = t[1] * l;
  if (e.shapeInfo.isUniform)
    return `
      float ${s}(int row, int col, int depth,
                  int depth2, int depth3, int depth4) {
        int index = round(dot(
          vec4(row, col, depth, depth2),
          vec4(${c}, ${l}, ${u}, ${o})) +
          dot(
            vec2(depth3, depth4),
            vec2(${i}, 1)));
        ${uu(e)}
      }
    `;
  let p = e.shapeInfo.flatOffset, d = e.shapeInfo.texShape, h = d[0], f = d[1];
  if (f === c && p == null)
    return `
      float ${s}(int row, int col, int depth,
                    int depth2, int depth3, int depth4) {
        int texR = row;
        float texC = dot(vec4(col, depth, depth2, depth3),
          vec4(${l}, ${u}, ${o}, ${i})) +
               float(depth4);
        vec2 uv = (vec2(texC, texR) + halfCR) /
                   vec2(${f}.0, ${h}.0);
        return sampleTexture(${n}, uv);
      }
    `;
  if (f === i && p == null)
    return `
      float ${s}(int row, int col, int depth,
                    int depth2, int depth3, int depth4) {
        float texR = dot(vec4(row, col, depth, depth2),
          vec4(${t[1] * t[2] * t[3] * t[4]},
               ${t[2] * t[3] * t[4]},
               ${t[3] * t[4]},
               ${t[4]})) + float(depth3);
        int texC = depth4;
        vec2 uv = (vec2(texC, texR) + halfCR) /
                  vec2(${f}.0, ${h}.0);
        return sampleTexture(${n}, uv);
      }
    `;
  let m = ki(n);
  return `
    float ${s}(int row, int col, int depth,
                  int depth2, int depth3, int depth4) {
      // Explicitly use integer operations as dot() only works on floats.
      int index = row * ${c} + col * ${l} + depth * ${u} +
          depth2 * ${o} + depth3 * ${i} + depth4 + ${m};
      vec2 uv = uvFromFlat(${h}, ${f}, index);
      return sampleTexture(${n}, uv);
    }
  `;
}
function uu(e) {
  let t = e.name, n = w.sizeFromShape(e.shapeInfo.logicalShape);
  return n < 2 ? `return ${t};` : `
    for (int i = 0; i < ${n}; i++) {
      if (i == index) {
        return ${t}[i];
      }
    }
  `;
}
function UK(e, t) {
  let n = e.name, s = n.charAt(0).toUpperCase() + n.slice(1), r = "get" + s + "AtOutCoords", a = e.shapeInfo.logicalShape.length, i = t.logicalShape.length, o = k1(e.shapeInfo.logicalShape, t.logicalShape), u = ot(i), l = i - a, c, p = ["x", "y", "z", "w", "u", "v"];
  a === 0 ? c = "" : i < 2 && o.length >= 1 ? c = "coords = 0;" : c = o.map((y) => `coords.${p[y + l]} = 0;`).join(`
`);
  let d = "";
  i < 2 && a > 0 ? d = "coords" : d = e.shapeInfo.logicalShape.map((y, v) => `coords.${p[v + l]}`).join(", ");
  let h = "return outputValue;", m = w.sizeFromShape(e.shapeInfo.logicalShape) === 1, b = w.sizeFromShape(t.logicalShape) === 1;
  if (a === 1 && !m && !b)
    h = `
      return vec4(outputValue.xy, outputValue.xy);
    `;
  else if (m && !b)
    i === 1 ? h = `
        return vec4(outputValue.x, outputValue.x, 0., 0.);
      ` : h = `
        return vec4(outputValue.x);
      `;
  else if (o.length) {
    let y = a - 2, v = a - 1;
    o.indexOf(y) > -1 && o.indexOf(v) > -1 ? h = "return vec4(outputValue.x);" : o.indexOf(y) > -1 ? h = "return vec4(outputValue.x, outputValue.y, outputValue.x, outputValue.y);" : o.indexOf(v) > -1 && (h = "return vec4(outputValue.xx, outputValue.zz);");
  }
  return `
    vec4 ${r}() {
      ${u} coords = getOutputCoords();
      ${c}
      vec4 outputValue = get${s}(${d});
      ${h}
    }
  `;
}
function GK(e, t) {
  let n = e.name, s = n.charAt(0).toUpperCase() + n.slice(1), r = "get" + s + "AtOutCoords", a = t.texShape, i = e.shapeInfo.texShape, o = e.shapeInfo.logicalShape.length, u = t.logicalShape.length;
  if (!e.shapeInfo.isUniform && o === u && e.shapeInfo.flatOffset == null && w.arraysEqual(i, a))
    return `
      float ${r}() {
        return sampleTexture(${n}, resultUV);
      }
    `;
  let l = ot(u), c = k1(e.shapeInfo.logicalShape, t.logicalShape), p = u - o, d, h = ["x", "y", "z", "w", "u", "v"];
  o === 0 ? d = "" : u < 2 && c.length >= 1 ? d = "coords = 0;" : d = c.map((m) => `coords.${h[m + p]} = 0;`).join(`
`);
  let f = "";
  return u < 2 && o > 0 ? f = "coords" : f = e.shapeInfo.logicalShape.map((m, g) => `coords.${h[g + p]}`).join(", "), `
    float ${r}() {
      ${l} coords = getOutputCoords();
      ${d}
      return get${s}(${f});
    }
  `;
}
function ot(e) {
  if (e <= 1)
    return "int";
  if (e === 2)
    return "ivec2";
  if (e === 3)
    return "ivec3";
  if (e === 4)
    return "ivec4";
  if (e === 5)
    return "ivec5";
  if (e === 6)
    return "ivec6";
  throw Error(`GPU for rank ${e} is not yet supported`);
}
function kv(e, t, n) {
  let { newShape: s, keptDims: r } = w.squeezeShape(t), a = t.length, i = e && a === 3 && t[0] === 1, o = i ? t.slice(1) : s, u = !e && a > 1 && !w.arraysEqual(t, n) && s.length < a || i;
  return { useSqueezeShape: u, uniformShape: u ? o : t, keptDims: r };
}
function lu(e, t) {
  let n = JSON.parse(JSON.stringify(e));
  return n.shapeInfo.logicalShape = t, n;
}
function cu(e, t) {
  return t.map((n) => e[n]).join(", ");
}
function HK(e, t, n, s) {
  let r = n.map((c, p) => {
    let d = { logicalShape: c.shape, texShape: c.isUniform ? null : c.texData.texShape, isUniform: c.isUniform, isPacked: c.isUniform ? false : c.texData.isPacked, flatOffset: null };
    return c.texData != null && c.texData.slice != null && c.texData.slice.flatOffset > 0 && (d.flatOffset = c.texData.slice.flatOffset), { name: t.variableNames[p], shapeInfo: d };
  }), a = r.map((c) => c.shapeInfo), i = { logicalShape: s.shape, texShape: s.texData.texShape, isUniform: false, isPacked: s.texData.isPacked, flatOffset: null }, o = lK(r, i, t), u = e1(e.gl, o), l = e.createProgram(u);
  return K().get("ENGINE_COMPILE_ONLY") ? { program: t, fragmentShader: u, source: o, webGLProgram: l, inShapeInfos: a, outShapeInfo: i, uniformLocations: null, customUniformLocations: null, infLoc: null, nanLoc: null, inShapesLocations: null, inTexShapesLocations: null, outShapeLocation: null, outShapeStridesLocation: null, outTexShapeLocation: null } : { program: t, fragmentShader: u, source: o, webGLProgram: l, inShapeInfos: a, outShapeInfo: i, ...C1(e, t, l) };
}
function C1(e, t, n) {
  let s = {}, r = {}, a = {}, i = [], o, u, l, c = null, p = null;
  p = e.getUniformLocation(n, "NAN", false), K().getNumber("WEBGL_VERSION") === 1 && (c = e.getUniformLocation(n, "INFINITY", false));
  let d = false;
  for (let h = 0; h < t.variableNames.length; h++) {
    let f = t.variableNames[h];
    s[f] = e.getUniformLocation(n, f, d), s[`offset${f}`] = e.getUniformLocation(n, `offset${f}`, d), t.enableShapeUniforms && (r[`${f}Shape`] = e.getUniformLocation(n, `${f}Shape`, d), a[`${f}TexShape`] = e.getUniformLocation(n, `${f}TexShape`, d));
  }
  return t.enableShapeUniforms && (o = e.getUniformLocation(n, "outShape", d), l = e.getUniformLocation(n, "outShapeStrides", d), u = e.getUniformLocation(n, "outTexShape", d)), t.customUniforms && t.customUniforms.forEach((h, f) => {
    i[f] = e.getUniformLocation(n, h.name, d);
  }), { uniformLocations: s, customUniformLocations: i, infLoc: c, nanLoc: p, inShapesLocations: r, inTexShapesLocations: a, outShapeLocation: o, outShapeStridesLocation: l, outTexShapeLocation: u };
}
function cw(e, t) {
  if (e.length !== t.length)
    throw Error(`Binary was compiled with ${e.length} inputs, but was executed with ${t.length} inputs`);
  e.forEach((n, s) => {
    let r = n.logicalShape, a = t[s], i = a.shape;
    if (!w.arraysEqual(r, i))
      throw Error(`Binary was compiled with different shapes than the current args. Shapes ${r} and ${i} must match`);
    if (n.isUniform && a.isUniform)
      return;
    let o = n.texShape, u = a.isUniform ? null : a.texData.texShape;
    if (!w.arraysEqual(o, u))
      throw Error(`Binary was compiled with different texture shapes than the current args. Shape ${o} and ${u} must match`);
  });
}
function qK(e, t, n, s, r) {
  t.program.enableShapeUniforms || (cw(t.inShapeInfos, n), cw([t.outShapeInfo], [s]));
  let a = s.texData.texture, i = s.texData.texShape;
  s.texData.isPacked ? e.setOutputPackedMatrixTexture(a.texture, i[0], i[1]) : e.setOutputMatrixTexture(a.texture, i[0], i[1]), e.setProgram(t.webGLProgram), K().getNumber("WEBGL_VERSION") === 1 && t.infLoc !== null && e.gl.uniform1f(t.infLoc, 1 / 0), t.nanLoc !== null && e.gl.uniform1f(t.nanLoc, NaN), n.forEach((u, l) => {
    let c = t.program.variableNames[l], p = t.uniformLocations[c], d = t.uniformLocations[`offset${c}`], h = t.inShapesLocations[`${c}Shape`], f = t.inTexShapesLocations[`${c}TexShape`];
    if (h) {
      let { uniformShape: m } = kv(t.program.packedInputs, u.shape, u.texData.texShape);
      switch (m.length) {
        case 1:
          e.gl.uniform1iv(h, new Int32Array(m));
          break;
        case 2:
          e.gl.uniform2iv(h, new Int32Array(m));
          break;
        case 3:
          e.gl.uniform3iv(h, new Int32Array(m));
          break;
        case 4:
          e.gl.uniform4iv(h, new Int32Array(m));
          break;
        default:
          break;
      }
    }
    if (f && e.gl.uniform2i(f, u.texData.texShape[0], u.texData.texShape[1]), p != null) {
      if (u.isUniform) {
        if (w.sizeFromShape(u.shape) < 2)
          e.gl.uniform1f(p, u.uniformValues[0]);
        else {
          let m = u.uniformValues;
          m instanceof Float32Array || (m = new Float32Array(m)), e.gl.uniform1fv(p, m);
        }
        return;
      }
      u.texData.slice != null && d != null && e.gl.uniform1i(d, u.texData.slice.flatOffset), e.setInputMatrixTexture(u.texData.texture.texture, p, l);
    }
  });
  let o = t.outShapeLocation;
  if (o)
    switch (s.shape.length) {
      case 1:
        e.gl.uniform1iv(o, new Int32Array(s.shape));
        break;
      case 2:
        e.gl.uniform2iv(o, new Int32Array(s.shape));
        break;
      case 3:
        e.gl.uniform3iv(o, new Int32Array(s.shape));
        break;
      case 4:
        e.gl.uniform4iv(o, new Int32Array(s.shape));
        break;
      default:
        break;
    }
  if (t.outShapeStridesLocation) {
    let u = w.computeStrides(s.shape);
    switch (s.shape.length) {
      case 2:
        e.gl.uniform1iv(t.outShapeStridesLocation, new Int32Array(u));
        break;
      case 3:
        e.gl.uniform2iv(t.outShapeStridesLocation, new Int32Array(u));
        break;
      case 4:
        e.gl.uniform3iv(t.outShapeStridesLocation, new Int32Array(u));
        break;
      default:
        break;
    }
  }
  t.outTexShapeLocation && e.gl.uniform2i(t.outTexShapeLocation, s.texData.texShape[0], s.texData.texShape[1]), t.program.customUniforms && r && t.program.customUniforms.forEach((u, l) => {
    let c = t.customUniformLocations[l], p = r[l];
    if (u.type === "float")
      e.gl.uniform1fv(c, p);
    else if (u.type === "vec2")
      e.gl.uniform2fv(c, p);
    else if (u.type === "vec3")
      e.gl.uniform3fv(c, p);
    else if (u.type === "vec4")
      e.gl.uniform4fv(c, p);
    else if (u.type === "int")
      e.gl.uniform1iv(c, p);
    else if (u.type === "ivec2")
      e.gl.uniform2iv(c, p);
    else if (u.type === "ivec3")
      e.gl.uniform3iv(c, p);
    else if (u.type === "ivec4")
      e.gl.uniform4iv(c, p);
    else
      throw Error(`uniform type ${u.type} is not supported yet.`);
  }), e.executeProgram();
}
function jK(e, t, n) {
  let s = "";
  t.concat(n).forEach((i) => {
    let o = i.texData != null && i.texData.slice != null && i.texData.slice.flatOffset > 0;
    if (e.enableShapeUniforms && !i.isUniform) {
      let u = i.texData.texShape, { useSqueezeShape: l, uniformShape: c, keptDims: p } = kv(e.packedInputs, i.shape, u), d = "", h = "", f = "";
      if (c.length === 1 && e.packedInputs) {
        let k = [Math.ceil(u[0] / 2), Math.ceil(u[1] / 2)];
        d = `${k[0] > 1}_${k[1] > 1}`;
      } else if (c.length === 2 && !e.packedInputs)
        h = `${c[0] > 1}_${c[1] > 1}`;
      else if (c.length > 2 && !e.packedInputs) {
        let k = w.computeStrides(c);
        f = `${k[0] === u[1]}_${k[k.length - 1] === u[1]}`;
      }
      let m = i.shape.length, g = c.length === 2 && w.arraysEqual(i.shape, u), b = w.sizeFromShape(i.shape) === 1, y = C.getBroadcastDims(i.shape, n.shape), v = !e.packedInputs && m === n.shape.length && w.arraysEqual(u, n.texData.texShape), x = e.packedInputs || c.length > 2 ? "" : `${u[0] > 1}_${u[1] > 1}`;
      s += `${m}_${v}_${l ? p : ""}_${c.length}_${b}_${y}_${g}_${d}_${h}_${f}_${x}_${o}`;
    } else {
      let u = i.isUniform ? "uniform" : i.texData.texShape;
      s += `${i.shape}_${u}_${o}`;
    }
  });
  let r = e.userCode, a = e.constructor.name;
  return a += "_" + s + "_" + r + `${K().getNumber("WEBGL_VERSION")}`, a;
}
function Sn(e) {
  return K().getBool("WEBGL_USE_SHAPES_UNIFORMS") && e <= 4;
}
var KK = class {
  constructor(e) {
    this.variableNames = ["A"], this.packedInputs = false, this.packedOutput = true, this.outPackingScheme = 0, this.customUniforms = [{ name: "texShape", type: "ivec2" }];
    let t = fn();
    this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length), this.userCode = `
      ivec3 outCoordsFromFlatIndex(int index) {
        ${this.enableShapeUniforms ? Jp(["r", "c", "d"], e) : wi(["r", "c", "d"], e)}
        return ivec3(r, c, d);
      }

      void main() {
        ivec2 resTexRC = ivec2(resultUV.yx * vec2(texShape[0], texShape[1]));
        int index = 4 * (resTexRC.x * texShape[1] + resTexRC.y);

        vec4 result = vec4(0.);

        for (int i=0; i<4; i++) {
          int flatIndex = index + i;
          ivec3 rc = outCoordsFromFlatIndex(flatIndex);
          result[i] = getA(rc.x, rc.y, rc.z);
        }

        ${t.output} = result;
      }
    `;
  }
};
var XK = class {
  constructor(e) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true, this.outPackingScheme = 0, this.customUniforms = [{ name: "texShape", type: "ivec2" }];
    let t = fn();
    this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length), this.userCode = `
      ivec3 outCoordsFromFlatIndex(int index) {
        ${this.enableShapeUniforms ? Jp(["r", "c", "d"], e) : wi(["r", "c", "d"], e)}
        return ivec3(r, c, d);
      }

      void main() {
        ivec2 resTexRC = ivec2(resultUV.yx * vec2(texShape[0], texShape[1]));
        int index = 4 * (resTexRC.x * texShape[1] + resTexRC.y);

        vec4 result = vec4(0.);

        for (int i=0; i<4; i++) {
          int flatIndex = index + i;
          ivec3 rc = outCoordsFromFlatIndex(flatIndex);
          result[i] = getChannel(getA(rc.x, rc.y, rc.z), vec2(rc.y, rc.z));
        }

        ${t.output} = result;
      }
    `;
  }
};
var YK = class {
  constructor(e) {
    this.variableNames = ["A"], this.outTexUsage = 3;
    let t = fn();
    this.outputShape = e, this.userCode = `
      ${w1}

      void main() {
        float x = getAAtOutCoords();
        ${t.output} = encode_float(x);
      }
    `;
  }
};
var QK = class {
  constructor(e) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = false, this.outTexUsage = 3;
    let t = fn();
    this.outputShape = e, this.userCode = `
      ${w1}

      void main() {
        ivec3 coords = getOutputCoords();
        float x = getChannel(getAAtOutCoords(), vec2(coords.y, coords.z));
        ${t.output} = encode_float(x);
      }
    `;
  }
};
var ZK = class {
  constructor(e, t = false) {
    this.variableNames = ["A"], this.customUniforms = [{ name: "texShape", type: "ivec2" }];
    let n = fn();
    this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length);
    let s = "result";
    t && (s = "floor(result * 255. + 0.5)"), this.userCode = `
      ${this.enableShapeUniforms ? wv() : xv(e)}

      void main() {
        ivec3 coords = getOutputCoords();

        int flatIndex = getFlatIndex(coords);
        int offset = imod(flatIndex, 4);

        flatIndex = idiv(flatIndex, 4, 1.);

        int r = flatIndex / texShape[1];
        int c = imod(flatIndex, texShape[1]);
        vec2 uv = (vec2(c, r) + halfCR) / vec2(texShape[1], texShape[0]);
        vec4 values = ${n.texture2D}(A, uv);

        float result;

        if(offset == 0) {
          result = values[0];
        } else if(offset == 1) {
          result = values[1];
        } else if(offset == 2) {
          result = values[2];
        } else {
          result = values[3];
        }

        ${n.output} = vec4(${s}, 0., 0., 0.);
      }
    `;
  }
};
var JK = class {
  constructor(e, t = false) {
    this.variableNames = ["A"], this.packedInputs = false, this.packedOutput = true, this.customUniforms = [{ name: "texShape", type: "ivec2" }];
    let n = fn();
    this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length);
    let s = "", r = "result";
    t && (r = "floor(result * 255. + 0.5)");
    for (let a = 0; a <= 1; a++)
      for (let i = 0; i <= 1; i++) {
        let o = a * 2 + i;
        s += `
          localCoords = coords;
          if(localCoords[2] + ${i} < ${this.enableShapeUniforms ? "outShape[2]" : `${e[2]}`}) {
          localCoords[2] += ${i};
          if (localCoords[1] + ${a} < ${this.enableShapeUniforms ? "outShape[1]" : `${e[1]}`}) {
            localCoords[1] += ${a};

            flatIndex = getFlatIndex(localCoords);
            offset = imod(flatIndex, 4);

            flatIndex = idiv(flatIndex, 4, 1.);

            int r = flatIndex / texShape[1];
            int c = imod(flatIndex, texShape[1]);
            vec2 uv = (vec2(c, r) + halfCR) / vec2(texShape[1], texShape[0]);
            values = ${n.texture2D}(A, uv);

            if (offset == 0) {
              result[${o}] = values[0];
            } else if (offset == 1) {
              result[${o}] = values[1];
            } else if (offset == 2) {
              result[${o}] = values[2];
            } else {
              result[${o}] = values[3];
            }
          }
        }
        `;
      }
    this.userCode = `
        ${this.enableShapeUniforms ? wv() : xv(e)}

        void main() {
          ivec3 coords = getOutputCoords();

          vec4 result = vec4(0.);
          int flatIndex, r, c, offset;
          ivec3 localCoords;
          vec2 uv;
          vec4 values;

          ${s}

          ${n.output} = ${r};
        }
    `;
  }
};
var eX = {};
Ee(eX, { bindVertexProgramAttributeStreams: () => F1, createBufferFromOutputTexture: () => z1, createFloat16MatrixTexture: () => A1, createFloat16PackedMatrixTexture: () => D1, createFloat32MatrixTexture: () => _1, createIndexBuffer: () => $1, createPackedMatrixTexture: () => R1, createUnsignedBytesMatrixTexture: () => E1, createVertexBuffer: () => T1, createVertexShader: () => N1, downloadByteEncodedFloatMatrixFromOutputTexture: () => L1, downloadFloat32MatrixFromBuffer: () => M1, downloadMatrixFromPackedOutputTexture: () => V1, downloadPackedMatrixFromBuffer: () => B1, getInternalFormatForFloat16MatrixTexture: () => Iv, getInternalFormatForFloat16PackedMatrixTexture: () => Tv, getInternalFormatForFloat32MatrixTexture: () => Sv, getInternalFormatForPackedMatrixTexture: () => Nv, getInternalFormatForUnsignedBytesMatrixTexture: () => Cv, uploadDenseMatrixToTexture: () => O1, uploadPixelDataToTexture: () => P1 });
function N1(e) {
  let t = fn(), n = `${t.version}
    precision highp float;
    ${t.attribute} vec3 clipSpacePos;
    ${t.attribute} vec2 uv;
    ${t.varyingVs} vec2 resultUV;

    void main() {
      gl_Position = vec4(clipSpacePos, 1);
      resultUV = uv;
    }`;
  return JC(e, n);
}
function T1(e) {
  let t = new Float32Array([-1, 1, 0, 0, 1, -1, -1, 0, 0, 0, 1, 1, 0, 1, 1, 1, -1, 0, 1, 0]);
  return s1(e, t);
}
function $1(e) {
  let t = new Uint16Array([0, 1, 2, 2, 1, 3]);
  return r1(e, t);
}
function Jl(e, t, n, s, r, a) {
  i1(t, n);
  let i = a1(e), o = e.TEXTURE_2D;
  return fe(e, () => e.bindTexture(o, i)), fe(e, () => e.texParameteri(o, e.TEXTURE_WRAP_S, e.CLAMP_TO_EDGE)), fe(e, () => e.texParameteri(o, e.TEXTURE_WRAP_T, e.CLAMP_TO_EDGE)), fe(e, () => e.texParameteri(o, e.TEXTURE_MIN_FILTER, e.NEAREST)), fe(e, () => e.texParameteri(o, e.TEXTURE_MAG_FILTER, e.NEAREST)), K().getNumber("WEBGL_VERSION") === 1 ? fe(e, () => e.texImage2D(o, 0, s, t, n, 0, r, a, null)) : fe(e, () => e.texStorage2D(o, 1, s, t, n)), fe(e, () => e.bindTexture(e.TEXTURE_2D, null)), { texture: i, texShape: [n, t] };
}
function Sv(e) {
  return e.internalFormatFloat;
}
function _1(e, t, n, s) {
  let [r, a] = Zl(t, n);
  return Jl(e, r, a, Sv(s), s.textureFormatFloat, e.FLOAT);
}
function Iv(e) {
  return e.internalFormatHalfFloat;
}
function A1(e, t, n, s) {
  let [r, a] = Zl(t, n);
  return Jl(e, r, a, Iv(s), s.textureFormatFloat, s.textureTypeHalfFloat);
}
function Cv(e) {
  return e.downloadTextureFormat;
}
function E1(e, t, n, s) {
  let [r, a] = Zl(t, n);
  return Jl(e, r, a, Cv(s), e.RGBA, e.UNSIGNED_BYTE);
}
function Nv(e) {
  return e.internalFormatPackedFloat;
}
function R1(e, t, n, s) {
  let [r, a] = au(t, n);
  return Jl(e, r, a, Nv(s), e.RGBA, e.FLOAT);
}
function Tv(e) {
  return e.internalFormatPackedHalfFloat;
}
function D1(e, t, n, s) {
  let [r, a] = au(t, n);
  return Jl(e, r, a, Tv(s), e.RGBA, s.textureTypeHalfFloat);
}
function F1(e, t, n) {
  return fe(e, () => e.bindBuffer(e.ARRAY_BUFFER, n)), jm(e, t, "clipSpacePos", n, 3, 20, 0) && jm(e, t, "uv", n, 2, 20, 12);
}
function O1(e, t, n, s, r, a) {
  fe(e, () => e.bindTexture(e.TEXTURE_2D, t));
  let i, o, u;
  r instanceof Uint8Array ? (i = new Uint8Array(n * s * 4), o = e.UNSIGNED_BYTE, u = e.RGBA) : (i = new Float32Array(n * s * 4), o = e.FLOAT, u = a.internalFormatPackedFloat), i.set(r), K().getNumber("WEBGL_VERSION") === 2 ? fe(e, () => e.texSubImage2D(e.TEXTURE_2D, 0, 0, 0, n, s, e.RGBA, o, i)) : fe(e, () => e.texImage2D(e.TEXTURE_2D, 0, u, n, s, 0, e.RGBA, o, i)), fe(e, () => e.bindTexture(e.TEXTURE_2D, null));
}
function P1(e, t, n) {
  fe(e, () => e.bindTexture(e.TEXTURE_2D, t)), n.data instanceof Uint8Array ? K().getNumber("WEBGL_VERSION") === 2 ? fe(e, () => e.texSubImage2D(e.TEXTURE_2D, 0, 0, 0, n.width, n.height, e.RGBA, e.UNSIGNED_BYTE, n.data)) : fe(e, () => e.texImage2D(e.TEXTURE_2D, 0, e.RGBA, n.width, n.height, 0, e.RGBA, e.UNSIGNED_BYTE, n.data)) : K().getNumber("WEBGL_VERSION") === 2 ? fe(e, () => e.texSubImage2D(e.TEXTURE_2D, 0, 0, 0, e.RGBA, e.UNSIGNED_BYTE, n)) : fe(e, () => e.texImage2D(e.TEXTURE_2D, 0, e.RGBA, e.RGBA, e.UNSIGNED_BYTE, n)), fe(e, () => e.bindTexture(e.TEXTURE_2D, null));
}
function z1(e, t, n, s) {
  let r = e.createBuffer();
  fe(e, () => e.bindBuffer(e.PIXEL_PACK_BUFFER, r));
  let o = 4 * 4 * t * n;
  return fe(e, () => e.bufferData(e.PIXEL_PACK_BUFFER, o, e.STREAM_READ)), fe(e, () => e.readPixels(0, 0, n, t, e.RGBA, e.FLOAT, 0)), fe(e, () => e.bindBuffer(e.PIXEL_PACK_BUFFER, null)), r;
}
function M1(e, t, n) {
  let s = e, r = new Float32Array(n);
  return s.bindBuffer(s.PIXEL_PACK_BUFFER, t), s.getBufferSubData(s.PIXEL_PACK_BUFFER, 0, r), s.bindBuffer(s.PIXEL_PACK_BUFFER, null), r;
}
function L1(e, t, n, s) {
  let [r, a] = Zl(t, n), i = 4, o = new Uint8Array(X5(t * n, i));
  return fe(e, () => e.readPixels(0, 0, r, a, s.downloadTextureFormat, e.UNSIGNED_BYTE, o)), new Float32Array(o.buffer);
}
function B1(e, t, n, s, r, a, i, o) {
  let u = e, l = new Float32Array(Y5(a, i));
  return u.bindBuffer(u.PIXEL_PACK_BUFFER, t), u.getBufferSubData(u.PIXEL_PACK_BUFFER, 0, l), u.bindBuffer(u.PIXEL_PACK_BUFFER, null), l;
}
function V1(e, t, n) {
  let s = new Float32Array(t * n * 4);
  return fe(e, () => e.readPixels(0, 0, n, t, e.RGBA, e.FLOAT, s)), s;
}
var em = class {
  constructor(e) {
    this.outputTexture = null, this.program = null, this.disposed = false, this.vertexAttrsAreBound = false, this.itemsToPoll = [];
    let t = K().getNumber("WEBGL_VERSION");
    e != null ? (this.gl = e, q5(t, e)) : this.gl = xs(t);
    let n = "WEBGL_color_buffer_float", s = "EXT_color_buffer_half_float";
    if (this.parallelCompilationExtension = this.gl.getExtension("KHR_parallel_shader_compile"), K().getNumber("WEBGL_VERSION") === 1) {
      let r = "OES_texture_float", a = "OES_texture_half_float";
      if (this.textureFloatExtension = Mu(this.gl, r), Ln(this.gl, a))
        this.textureHalfFloatExtension = Mu(this.gl, a);
      else if (K().get("WEBGL_FORCE_F16_TEXTURES"))
        throw new Error("GL context does not support half float textures, yet the environment flag WEBGL_FORCE_F16_TEXTURES is set to true.");
      if (this.colorBufferFloatExtension = this.gl.getExtension(n), Ln(this.gl, s))
        this.colorBufferHalfFloatExtension = Mu(this.gl, s);
      else if (K().get("WEBGL_FORCE_F16_TEXTURES"))
        throw new Error("GL context does not support color renderable half floats, yet the environment flag WEBGL_FORCE_F16_TEXTURES is set to true.");
    } else if (n = "EXT_color_buffer_float", Ln(this.gl, n))
      this.colorBufferFloatExtension = this.gl.getExtension(n);
    else if (Ln(this.gl, s))
      this.colorBufferHalfFloatExtension = this.gl.getExtension(s);
    else
      throw new Error("GL context does not support color renderable floats");
    this.vertexBuffer = T1(this.gl), this.indexBuffer = $1(this.gl), this.framebuffer = o1(this.gl), this.textureConfig = yv(this.gl, this.textureHalfFloatExtension);
  }
  get debug() {
    return K().getBool("DEBUG");
  }
  dispose() {
    if (this.disposed)
      return;
    this.program != null && console.warn("Disposing a GPGPUContext that still has a bound WebGLProgram. This is probably a resource leak, delete the program with GPGPUContext.deleteProgram before disposing."), this.outputTexture != null && console.warn("Disposing a GPGPUContext that still has a bound output matrix texture.  This is probably a resource leak, delete the output matrix texture with GPGPUContext.deleteMatrixTexture before disposing.");
    let e = this.gl;
    fe(e, () => e.finish()), fe(e, () => e.bindFramebuffer(e.FRAMEBUFFER, null)), fe(e, () => e.deleteFramebuffer(this.framebuffer)), fe(e, () => e.bindBuffer(e.ARRAY_BUFFER, null)), fe(e, () => e.bindBuffer(e.ELEMENT_ARRAY_BUFFER, null)), fe(e, () => e.deleteBuffer(this.indexBuffer)), this.disposed = true;
  }
  createFloat32MatrixTexture(e, t) {
    return this.throwIfDisposed(), _1(this.gl, e, t, this.textureConfig);
  }
  createFloat16MatrixTexture(e, t) {
    return this.throwIfDisposed(), A1(this.gl, e, t, this.textureConfig);
  }
  createUnsignedBytesMatrixTexture(e, t) {
    return this.throwIfDisposed(), E1(this.gl, e, t, this.textureConfig);
  }
  uploadPixelDataToTexture(e, t) {
    this.throwIfDisposed(), P1(this.gl, e, t);
  }
  uploadDenseMatrixToTexture(e, t, n, s) {
    this.throwIfDisposed(), O1(this.gl, e, t, n, s, this.textureConfig);
  }
  createFloat16PackedMatrixTexture(e, t) {
    return this.throwIfDisposed(), D1(this.gl, e, t, this.textureConfig);
  }
  createPackedMatrixTexture(e, t) {
    return this.throwIfDisposed(), R1(this.gl, e, t, this.textureConfig);
  }
  deleteMatrixTexture(e) {
    this.throwIfDisposed(), this.outputTexture === e && (Km(this.gl, this.framebuffer), this.outputTexture = null), fe(this.gl, () => this.gl.deleteTexture(e));
  }
  downloadByteEncodedFloatMatrixFromOutputTexture(e, t, n) {
    return this.downloadMatrixDriver(e, () => L1(this.gl, t, n, this.textureConfig));
  }
  downloadPackedMatrixFromBuffer(e, t, n, s, r, a) {
    return B1(this.gl, e, t, n, s, r, a, this.textureConfig);
  }
  downloadFloat32MatrixFromBuffer(e, t) {
    return M1(this.gl, e, t);
  }
  createBufferFromTexture(e, t, n) {
    this.bindTextureToFrameBuffer(e);
    let s = z1(this.gl, t, n, this.textureConfig);
    return this.unbindTextureToFrameBuffer(), s;
  }
  createAndWaitForFence() {
    let e = this.createFence(this.gl);
    return this.pollFence(e);
  }
  createFence(e) {
    let t, n;
    if (K().getBool("WEBGL_FENCE_API_ENABLED")) {
      let s = e, r = s.fenceSync(s.SYNC_GPU_COMMANDS_COMPLETE, 0);
      e.flush(), n = () => {
        let a = s.clientWaitSync(r, 0, 0);
        return a === s.ALREADY_SIGNALED || a === s.CONDITION_SATISFIED;
      }, t = r;
    } else
      K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION") > 0 ? (t = this.beginQuery(), this.endQuery(), n = () => this.isQueryAvailable(t, K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION"))) : n = () => true;
    return { query: t, isFencePassed: n };
  }
  downloadMatrixFromPackedTexture(e, t, n) {
    return this.downloadMatrixDriver(e, () => V1(this.gl, t, n));
  }
  createProgram(e) {
    this.throwIfDisposed();
    let t = this.gl;
    this.vertexShader == null && (this.vertexShader = N1(t));
    let n = t1(t);
    return fe(t, () => t.attachShader(n, this.vertexShader)), fe(t, () => t.attachShader(n, e)), n1(t, n), this.debug && od(t, n), this.vertexAttrsAreBound || (this.setProgram(n), this.vertexAttrsAreBound = F1(t, this.program, this.vertexBuffer)), n;
  }
  deleteProgram(e) {
    this.throwIfDisposed(), e === this.program && (this.program = null), e != null && fe(this.gl, () => this.gl.deleteProgram(e));
  }
  setProgram(e) {
    this.throwIfDisposed(), this.program = e, this.program != null && this.debug && od(this.gl, this.program), fe(this.gl, () => this.gl.useProgram(e));
  }
  getUniformLocation(e, t, n = true) {
    return this.throwIfDisposed(), n ? l1(this.gl, e, t) : c1(this.gl, e, t);
  }
  getAttributeLocation(e, t) {
    return this.throwIfDisposed(), fe(this.gl, () => this.gl.getAttribLocation(e, t));
  }
  getUniformLocationNoThrow(e, t) {
    return this.throwIfDisposed(), this.gl.getUniformLocation(e, t);
  }
  setInputMatrixTexture(e, t, n) {
    this.throwIfDisposed(), this.throwIfNoProgram(), d1(this.gl, e, t, n);
  }
  setOutputMatrixTexture(e, t, n) {
    this.setOutputMatrixTextureDriver(e, n, t);
  }
  setOutputPackedMatrixTexture(e, t, n) {
    this.throwIfDisposed();
    let [s, r] = au(t, n);
    this.setOutputMatrixTextureDriver(e, s, r);
  }
  setOutputMatrixWriteRegion(e, t, n, s) {
    this.setOutputMatrixWriteRegionDriver(n, e, s, t);
  }
  setOutputPackedMatrixWriteRegion(e, t, n, s) {
    throw new Error("setOutputPackedMatrixWriteRegion not implemented.");
  }
  debugValidate() {
    this.program != null && od(this.gl, this.program), Lu(this.gl);
  }
  executeProgram() {
    this.throwIfDisposed(), this.throwIfNoProgram();
    let e = this.gl;
    this.debug && this.debugValidate(), fe(e, () => e.drawElements(e.TRIANGLES, 6, e.UNSIGNED_SHORT, 0));
  }
  blockUntilAllProgramsCompleted() {
    this.throwIfDisposed(), fe(this.gl, () => this.gl.finish());
  }
  getQueryTimerExtension() {
    return this.disjointQueryTimerExtension == null && (this.disjointQueryTimerExtension = Mu(this.gl, K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION") === 2 ? "EXT_disjoint_timer_query_webgl2" : "EXT_disjoint_timer_query")), this.disjointQueryTimerExtension;
  }
  getQueryTimerExtensionWebGL2() {
    return this.getQueryTimerExtension();
  }
  getQueryTimerExtensionWebGL1() {
    return this.getQueryTimerExtension();
  }
  beginQuery() {
    if (K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION") === 2) {
      let n = this.gl, s = this.getQueryTimerExtensionWebGL2(), r = n.createQuery();
      return n.beginQuery(s.TIME_ELAPSED_EXT, r), r;
    }
    let e = this.getQueryTimerExtensionWebGL1(), t = e.createQueryEXT();
    return e.beginQueryEXT(e.TIME_ELAPSED_EXT, t), t;
  }
  endQuery() {
    if (K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION") === 2) {
      let t = this.gl, n = this.getQueryTimerExtensionWebGL2();
      t.endQuery(n.TIME_ELAPSED_EXT);
      return;
    }
    let e = this.getQueryTimerExtensionWebGL1();
    e.endQueryEXT(e.TIME_ELAPSED_EXT);
  }
  async waitForQueryAndGetTime(e) {
    return await w.repeatedTry(() => this.disposed || this.isQueryAvailable(e, K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION"))), this.getQueryTime(e, K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_VERSION"));
  }
  getQueryTime(e, t) {
    if (t === 0)
      return null;
    if (t === 2) {
      let n = this.gl;
      return n.getQueryParameter(e, n.QUERY_RESULT) / 1e6;
    } else {
      let n = this.getQueryTimerExtensionWebGL1();
      return n.getQueryObjectEXT(e, n.QUERY_RESULT_EXT) / 1e6;
    }
  }
  isQueryAvailable(e, t) {
    if (t === 0)
      return true;
    if (t === 2) {
      let n = this.gl, s = this.getQueryTimerExtensionWebGL2(), r = n.getQueryParameter(e, n.QUERY_RESULT_AVAILABLE);
      return this.disjoint == null && (this.disjoint = this.gl.getParameter(s.GPU_DISJOINT_EXT)), r && !this.disjoint;
    } else {
      let n = this.getQueryTimerExtensionWebGL1(), s = n.getQueryObjectEXT(e, n.QUERY_RESULT_AVAILABLE_EXT);
      return this.disjoint == null && (this.disjoint = this.gl.getParameter(n.GPU_DISJOINT_EXT)), s && !this.disjoint;
    }
  }
  pollFence(e) {
    return new Promise((t) => {
      this.addItemToPoll(() => e.isFencePassed(), () => t());
    });
  }
  pollItems() {
    let e = tX(this.itemsToPoll.map((t) => t.isDoneFn));
    for (let t = 0; t <= e; ++t) {
      let { resolveFn: n } = this.itemsToPoll[t];
      n();
    }
    this.itemsToPoll = this.itemsToPoll.slice(e + 1);
  }
  addItemToPoll(e, t) {
    this.itemsToPoll.push({ isDoneFn: e, resolveFn: t }), !(this.itemsToPoll.length > 1) && w.repeatedTry(() => (this.pollItems(), this.itemsToPoll.length === 0));
  }
  bindTextureToFrameBuffer(e) {
    this.throwIfDisposed(), ud(this.gl, e, this.framebuffer), this.debug && Lu(this.gl);
  }
  unbindTextureToFrameBuffer() {
    this.outputTexture != null ? (ud(this.gl, this.outputTexture, this.framebuffer), this.debug && Lu(this.gl)) : Km(this.gl, this.framebuffer);
  }
  downloadMatrixDriver(e, t) {
    this.bindTextureToFrameBuffer(e);
    let n = t();
    return this.unbindTextureToFrameBuffer(), n;
  }
  setOutputMatrixTextureDriver(e, t, n) {
    this.throwIfDisposed();
    let s = this.gl;
    ud(s, e, this.framebuffer), this.debug && Lu(s), this.outputTexture = e, fe(s, () => s.viewport(0, 0, t, n)), fe(s, () => s.scissor(0, 0, t, n));
  }
  setOutputMatrixWriteRegionDriver(e, t, n, s) {
    this.throwIfDisposed(), fe(this.gl, () => this.gl.scissor(e, t, n, s));
  }
  throwIfDisposed() {
    if (this.disposed)
      throw new Error("Attempted to use disposed GPGPUContext.");
  }
  throwIfNoProgram() {
    if (this.program == null)
      throw new Error("No GPU program is currently set.");
  }
};
function tX(e) {
  let t = 0;
  for (; t < e.length && e[t](); ++t)
    ;
  return t - 1;
}
var { addImpl: nX, bincountImpl: W1, bincountReduceImpl: sX, ceilImpl: rX, concatImpl: aX, equalImpl: iX, expImpl: oX, expm1Impl: uX, floorImpl: lX, gatherNdImpl: cX, gatherV2Impl: dX, greaterImpl: pX, greaterEqualImpl: hX, lessImpl: fX, lessEqualImpl: mX, linSpaceImpl: gX, logImpl: bX, maxImpl: yX, maximumImpl: vX, minimumImpl: xX, multiplyImpl: wX, negImpl: kX, notEqualImpl: SX, prodImpl: IX, rangeImpl: CX, rsqrtImpl: NX, scatterImpl: TX, sigmoidImpl: $X, simpleAbsImpl: U1, sliceImpl: _X, sparseFillEmptyRowsImpl: AX, sparseReshapeImpl: EX, sparseSegmentReductionImpl: G1, sqrtImpl: RX, stridedSliceImpl: DX, stringNGramsImpl: FX, stringSplitImpl: OX, stringToHashBucketFastImpl: PX, subImpl: zX, tileImpl: MX, topKImpl: LX, transposeImpl: $v, uniqueImpl: BX } = iv;
function H1(e, t) {
  return ["x", "y", "z", "w", "u", "v"].slice(0, t).map((n) => `${e}.${n}`);
}
function ln(e, t) {
  return t === 1 ? [e] : H1(e, t);
}
function VX(e, t) {
  if (e === 1)
    return "rc";
  let n = "";
  for (let s = 0; s < e; s++)
    n += t[s], s < e - 1 && (n += ",");
  return n;
}
var WX = class {
  constructor(e) {
    if (this.variableNames = ["A"], this.packedInputs = false, this.packedOutput = true, this.outputShape = e, this.rank = e.length, this.enableShapeUniforms = Sn(this.outputShape.length), this.rank === 0)
      this.userCode = `
        void main() {
          setOutput(vec4(getA(), 0., 0., 0.));
        }
      `;
    else {
      let t = ln("rc", this.rank), n = ot(this.rank), s = this.getOutOfBoundsCondition(t), r = this.getSetup(t), a = this.getOutput(t);
      this.userCode = `
        void main() {
          ${n} rc = getOutputCoords();

          if(${s}) {
            setOutput(vec4(0));
          } else {
            ${r}

            setOutput(vec4(${a}));
          }
        }
      `;
    }
  }
  getSourceCoordsArr(e) {
    let t = [];
    for (let n = 0; n <= 1; n++)
      for (let s = 0; s <= 1; s++) {
        let r = `${n === 0 ? "r" : "rp1"}, ${s === 0 ? "c" : "cp1"}`;
        for (let a = 2; a < this.rank; a++)
          r = `${e[e.length - 1 - a]},` + r;
        t.push(r);
      }
    return t;
  }
  getOutOfBoundsCondition(e) {
    if (this.rank === 1)
      return `rc > ${this.enableShapeUniforms ? "outShape" : this.outputShape[0]}`;
    let t = "";
    for (let n = this.rank - 2; n < this.rank; n++)
      t += `${e[n]} >= ${this.enableShapeUniforms ? `outShape[${n}]` : this.outputShape[n]}`, n < this.rank - 1 && (t += "||");
    return t;
  }
  getSetup(e) {
    if (this.rank === 1)
      return "";
    let t = e.slice(-2), n = this.enableShapeUniforms ? `outShape[${this.rank} - 1]` : this.outputShape[this.rank - 1], s = this.enableShapeUniforms ? `outShape[${this.rank} - 2]` : this.outputShape[this.rank - 2];
    return `
      int r = ${t[0]};
      int c = ${t[1]};
      int rp1 = r + 1;
      int cp1 = c + 1;

      bool cEdge = cp1 >= ${n};
      bool rEdge = rp1 >= ${s};
    `;
  }
  getOutput(e) {
    let t = this.getSourceCoordsArr(e);
    return this.rank === 1 ? `getA(rc), (rc + 1 >= ${this.enableShapeUniforms ? "outShape" : this.outputShape[0]} ? 0. : getA(rc + 1)), 0, 0` : `getA(${t[0]}),
            cEdge ? 0. : getA(${t[1]}),
            rEdge ? 0. : getA(${t[2]}),
            rEdge || cEdge ? 0. : getA(${t[3]})`;
  }
};
var q1 = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true, this.customUniforms = [{ name: "inputShape", type: "ivec3" }], this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length);
    let n = "";
    for (let s = 0; s < 4; s++) {
      let r = "thisRC = rc;";
      s % 2 === 1 && (r += "thisRC.z += 1;"), s > 1 && (r += "thisRC.y += 1;"), n += `
        ${r}
        ${s > 0 ? "if(thisRC.y < rows && thisRC.z < cols){" : ""}
          int flatIndex = getFlatIndex(thisRC);

          ivec3 inputRC = inputCoordsFromReshapedOutCoords(flatIndex);
          vec2 inputRCInnerDims = vec2(float(inputRC.y),float(inputRC.z));

          result[${s}] =
            getChannel(getA(inputRC.x, inputRC.y, inputRC.z), inputRCInnerDims);
        ${s > 0 ? "}" : ""}
      `;
    }
    this.userCode = `
      ${UX(t, this.enableShapeUniforms)}
      ${this.enableShapeUniforms ? wv() : xv(e)}

      void main() {
        ivec3 rc = getOutputCoords();

        vec4 result = vec4(0.);

        ivec3 thisRC;
        int rows = ${this.enableShapeUniforms ? "outShape[1]" : e[1]};
        int cols = ${this.enableShapeUniforms ? "outShape[2]" : e[2]};

        ${n}

        setOutput(result);
      }
    `;
  }
};
function UX(e, t) {
  return `
    ivec3 inputCoordsFromReshapedOutCoords(int index) {
      ${t ? uK(["r", "c", "d"], "inputShape") : wi(["r", "c", "d"], e)}
      return ivec3(r, c, d);
    }
  `;
}
var GX = class {
  constructor(e) {
    this.gpgpu = e, this.numUsedTextures = 0, this.numFreeTextures = 0, this._numBytesAllocated = 0, this._numBytesFree = 0, this.freeTextures = {}, this.logEnabled = false, this.usedTextures = {};
  }
  acquireTexture(e, t, n) {
    let s = pw(t, n), r = hw(e, s, n);
    r in this.freeTextures || (this.freeTextures[r] = []), r in this.usedTextures || (this.usedTextures[r] = []);
    let a = dw(e, s, this.gpgpu.gl, this.gpgpu.textureConfig, n);
    if (this.freeTextures[r].length > 0) {
      this.numFreeTextures--, this.numUsedTextures++, this._numBytesFree -= a, this.log();
      let o = this.freeTextures[r].shift();
      return this.usedTextures[r].push(o), o;
    }
    let i;
    return s === 3 ? i = this.gpgpu.createPackedMatrixTexture(e[0], e[1]) : s === 4 ? i = this.gpgpu.createFloat16PackedMatrixTexture(e[0], e[1]) : s === 1 ? i = this.gpgpu.createFloat32MatrixTexture(e[0], e[1]) : s === 0 ? i = this.gpgpu.createFloat16MatrixTexture(e[0], e[1]) : s === 2 && (i = this.gpgpu.createUnsignedBytesMatrixTexture(e[0], e[1])), this.usedTextures[r].push(i), this.numUsedTextures++, this._numBytesAllocated += a, this.log(), i;
  }
  releaseTexture(e, t, n, s) {
    if (this.freeTextures == null)
      return;
    let r = pw(n, s), a = hw(t, r, s);
    a in this.freeTextures || (this.freeTextures[a] = []);
    let i = dw(t, r, this.gpgpu.gl, this.gpgpu.textureConfig, s), o = K().get("WEBGL_DELETE_TEXTURE_THRESHOLD");
    o !== -1 && this._numBytesAllocated > o ? (this.gpgpu.deleteMatrixTexture(e.texture), this._numBytesAllocated -= i) : (this.freeTextures[a].push(e), this.numFreeTextures++, this._numBytesFree += i), this.numUsedTextures--;
    let u = this.usedTextures[a], l = u.indexOf(e);
    if (l < 0)
      throw new Error("Cannot release a texture that was never provided by this texture manager");
    u.splice(l, 1), this.log();
  }
  log() {
    if (!this.logEnabled)
      return;
    let e = this.numFreeTextures + this.numUsedTextures;
    console.log("Free/Used", `${this.numFreeTextures} / ${this.numUsedTextures}`, `(${e})`);
    let t = this._numBytesFree / this._numBytesAllocated;
    console.log(`Bytes allocated: ${this._numBytesAllocated}`), console.log(`Bytes unused: ${this._numBytesFree} (${Math.round(100 * t)}%)`);
  }
  get numBytesAllocated() {
    return this._numBytesAllocated;
  }
  get numBytesFree() {
    return this._numBytesFree;
  }
  getNumUsedTextures() {
    return this.numUsedTextures;
  }
  getNumFreeTextures() {
    return this.numFreeTextures;
  }
  dispose() {
    if (this.freeTextures != null) {
      for (let e in this.freeTextures)
        this.freeTextures[e].forEach((t) => {
          this.gpgpu.deleteMatrixTexture(t.texture);
        });
      for (let e in this.usedTextures)
        this.usedTextures[e].forEach((t) => {
          this.gpgpu.deleteMatrixTexture(t.texture);
        });
      this.freeTextures = null, this.usedTextures = null, this.numUsedTextures = 0, this.numFreeTextures = 0, this._numBytesAllocated = 0, this._numBytesFree = 0;
    }
  }
};
function HX(e, t) {
  let n = e;
  if (t === n.R32F)
    return 4;
  if (t === n.R16F)
    return 2;
  if (t === n.RGBA32F)
    return 16;
  if (t === e.RGBA)
    return 16;
  if (t === n.RGBA16F)
    return 8;
  if (t === n.RGBA8)
    return 4;
  throw new Error(`Unknown internal format ${t}`);
}
function dw(e, t, n, s, r) {
  let a = qX(t, s), i;
  if (r) {
    let [u, l] = au(e[0], e[1]);
    i = u * l;
  } else {
    let [u, l] = Zl(e[0], e[1]);
    i = u * l;
  }
  let o = HX(n, a);
  return i * o;
}
function qX(e, t) {
  switch (e) {
    case 3:
      return Nv(t);
    case 4:
      return Tv(t);
    case 1:
      return Sv(t);
    case 0:
      return Iv(t);
    case 2:
      return Cv(t);
    default:
      throw new Error(`Unknown physical texture type ${e}`);
  }
}
function jX(e) {
  return K().getBool("WEBGL_RENDER_FLOAT32_ENABLED") ? e ? 3 : 1 : e ? 4 : 0;
}
function pw(e, t) {
  if (e === 1)
    return 3;
  if (e === 0 || e == null)
    return jX(t);
  if (e === 3 || e === 2)
    return 2;
  throw new Error(`Unknown logical texture type ${e}`);
}
function hw(e, t, n) {
  return `${e[0]}_${e[1]}_${t}_${n}`;
}
var Gs = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length), this.userCode = `
      float unaryOperation(float x) {
        ${t}
      }

      void main() {
        float x = getAAtOutCoords();
        float y = unaryOperation(x);

        setOutput(y);
      }
    `;
  }
};
var ss = "if (isnan(x)) return x;";
var KX = "return x;";
var fw = "return abs(x);";
var XX = "return (x >= 0.0) ? x : (exp(x) - 1.0);";
var YX = ss + `
  return (x < 0.0) ? 0.0 : x;
`;
var QX = ss + `
  return (x < 0.0) ? 0.0 : min(6.0, x);
`;
var Vi = "return x;";
var ZX = "return 1.0 / (1.0 + exp(-1.0 * x));";
var JX = "return x;";
var e8 = `
  vec4 result;

  result.r = (x.r >= 0.0) ? x.r : (exp(x.r) - 1.0);
  result.g = (x.g >= 0.0) ? x.g : (exp(x.g) - 1.0);
  result.b = (x.b >= 0.0) ? x.b : (exp(x.b) - 1.0);
  result.a = (x.a >= 0.0) ? x.a : (exp(x.a) - 1.0);

  return result;
`;
var t8 = `
  vec4 result = x * vec4(greaterThanEqual(x, vec4(0.0)));
  bvec4 isNaN = isnan(x);

  result.r = isNaN.r ? x.r : result.r;
  result.g = isNaN.g ? x.g : result.g;
  result.b = isNaN.b ? x.b : result.b;
  result.a = isNaN.a ? x.a : result.a;

  return result;
`;
var n8 = `
  vec4 result = min(x, vec4(6.)) * vec4(greaterThanEqual(x, vec4(0.0)));
  bvec4 isNaN = isnan(x);

  result.r = isNaN.r ? x.r : result.r;
  result.g = isNaN.g ? x.g : result.g;
  result.b = isNaN.b ? x.b : result.b;
  result.a = isNaN.a ? x.a : result.a;

  return result;
`;
var s8 = "return 1.0 / (1.0 + exp(-1.0 * x));";
var Jr = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true, this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length), this.userCode = `
      vec4 unaryOperation(vec4 x) {
        ${t}
      }

      void main() {
        vec4 x = getAAtOutCoords();
        vec4 y = unaryOperation(x);

        setOutput(y);
      }
    `;
  }
};
var r8 = class {
  constructor(e) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = false, this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length);
    let t = e.length, n = ln("rc", t), s = ot(t), r = VX(t, n), a = n.slice(-2), i = t <= 1 ? "rc" : `vec2(${a.join(",")})`;
    this.userCode = `
      void main() {
        ${s} rc = getOutputCoords();
        vec4 packedInput = getA(${r});

        setOutput(getChannel(packedInput, ${i}));
      }
    `;
  }
};
var a8 = ws.whereImpl;
var i8 = 1e-7;
var o8 = 1e-4;
var td = {};
function u8(e) {
  return e in td || (td[e] = {}), td[e];
}
var l8 = K().getNumber("CPU_HANDOFF_SIZE_THRESHOLD");
var c8 = 600;
function d8() {
  return K().global.screen == null ? 1024 : K().global.screen.height * K().global.screen.width * window.devicePixelRatio * c8 / 1024 / 1024;
}
var j1 = class extends il {
  constructor(e) {
    if (super(), this.pendingRead = /* @__PURE__ */ new WeakMap(), this.pendingDisposal = /* @__PURE__ */ new WeakSet(), this.dataRefCount = /* @__PURE__ */ new WeakMap(), this.numBytesInGPU = 0, this.uploadWaitMs = 0, this.downloadWaitMs = 0, this.lastGlFlushTime = 0, this.warnedAboutMemory = false, this.pendingDeletes = 0, this.disposed = false, !K().getBool("HAS_WEBGL"))
      throw new Error("WebGL is not supported on this device");
    let t;
    if (e != null) {
      if (e instanceof em)
        t = e;
      else {
        let n = xs(K().getNumber("WEBGL_VERSION"), e);
        t = new em(n);
      }
      this.binaryCache = {}, this.gpgpuCreatedLocally = false;
    } else {
      let n = xs(K().getNumber("WEBGL_VERSION"));
      t = new em(n), this.binaryCache = u8(K().getNumber("WEBGL_VERSION")), this.gpgpuCreatedLocally = true;
    }
    this.gpgpu = t, this.canvas = this.gpgpu.gl.canvas, this.textureManager = new GX(this.gpgpu), this.numMBBeforeWarning = d8(), this.texData = new Yd(this, ds());
  }
  nextDataId() {
    return j1.nextDataId++;
  }
  numDataIds() {
    return this.texData.numDataIds() - this.pendingDeletes;
  }
  write(e, t, n) {
    if ((K().getBool("WEBGL_CHECK_NUMERICAL_PROBLEMS") || K().getBool("DEBUG")) && this.checkNumericalProblems(e), n === "complex64" && e != null)
      throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");
    let s = { id: this.nextDataId() };
    return this.texData.set(s, { shape: t, dtype: n, values: e, usage: 1, refCount: 1 }), s;
  }
  refCount(e) {
    return this.texData.has(e) ? this.texData.get(e).refCount : 0;
  }
  incRef(e) {
    let t = this.texData.get(e);
    t.refCount++;
  }
  decRef(e) {
    if (this.texData.has(e)) {
      let t = this.texData.get(e);
      t.refCount--;
    }
  }
  move(e, t, n, s, r) {
    if (K().getBool("DEBUG") && this.checkNumericalProblems(t), s === "complex64")
      throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");
    this.texData.set(e, { shape: n, dtype: s, values: t, usage: 1, refCount: r });
  }
  disposeIntermediateTensorInfo(e) {
    this.disposeData(e.dataId);
  }
  readSync(e) {
    let t = this.texData.get(e), { values: n, dtype: s, complexTensorInfos: r, slice: a, shape: i, isPacked: o } = t;
    if (a != null) {
      let p;
      o ? p = new Jr(i, Vi) : p = new Gs(i, Vi);
      let d = this.runWebGLProgram(p, [{ dataId: e, shape: i, dtype: s }], s), h = this.readSync(d.dataId);
      return this.disposeIntermediateTensorInfo(d), h;
    }
    if (n != null)
      return this.convertAndCacheOnCPU(e);
    if (s === "string")
      return n;
    let u = this.activeTimers != null, l;
    u && (l = w.now());
    let c;
    if (s === "complex64") {
      let p = this.readSync(r.real.dataId), d = this.readSync(r.imag.dataId);
      c = C.mergeRealAndImagArrays(p, d);
    } else
      c = this.getValuesFromTexture(e);
    return u && (this.downloadWaitMs += w.now() - l), this.convertAndCacheOnCPU(e, c);
  }
  async read(e) {
    if (this.pendingRead.has(e)) {
      let h = this.pendingRead.get(e);
      return new Promise((f) => h.push(f));
    }
    let t = this.texData.get(e), { values: n, shape: s, slice: r, dtype: a, complexTensorInfos: i, isPacked: o } = t;
    if (r != null) {
      let h;
      o ? h = new Jr(s, Vi) : h = new Gs(s, Vi);
      let f = this.runWebGLProgram(h, [{ dataId: e, shape: s, dtype: a }], a), m = this.read(f.dataId);
      return this.disposeIntermediateTensorInfo(f), m;
    }
    if (n != null)
      return this.convertAndCacheOnCPU(e);
    if (K().getBool("DEBUG") && !K().getBool("WEBGL_DOWNLOAD_FLOAT_ENABLED") && K().getNumber("WEBGL_VERSION") === 2)
      throw new Error("tensor.data() with WEBGL_DOWNLOAD_FLOAT_ENABLED=false and WEBGL_VERSION=2 not yet supported.");
    let u = null, l;
    if (a !== "complex64" && K().get("WEBGL_BUFFER_SUPPORTED")) {
      l = this.decode(e);
      let h = this.texData.get(l.dataId);
      u = this.gpgpu.createBufferFromTexture(h.texture.texture, ...Jc(s));
    }
    this.pendingRead.set(e, []), a !== "complex64" && await this.gpgpu.createAndWaitForFence();
    let c;
    if (a === "complex64") {
      let h = await Promise.all([this.read(i.real.dataId), this.read(i.imag.dataId)]), f = h[0], m = h[1];
      c = C.mergeRealAndImagArrays(f, m);
    } else if (u == null)
      c = this.getValuesFromTexture(e);
    else {
      let h = w.sizeFromShape(s);
      c = this.gpgpu.downloadFloat32MatrixFromBuffer(u, h);
    }
    if (l != null && this.disposeIntermediateTensorInfo(l), u != null) {
      let h = this.gpgpu.gl;
      fe(h, () => h.deleteBuffer(u));
    }
    let p = this.convertAndCacheOnCPU(e, c), d = this.pendingRead.get(e);
    return this.pendingRead.delete(e), d.forEach((h) => h(p)), this.pendingDisposal.has(e) && (this.pendingDisposal.delete(e), this.disposeData(e) && ds().removeDataId(e, this), this.pendingDeletes--), p;
  }
  readToGPU(e, t = {}) {
    let n = this.texData.get(e), { values: s, shape: r, slice: a, dtype: i, isPacked: o, texture: u } = n;
    if (i === "complex64")
      throw new Error("Does not support reading texture for complex64 dtype.");
    if (a != null) {
      let d;
      o ? d = new Jr(r, Vi) : d = new Gs(r, Vi);
      let h = this.runWebGLProgram(d, [{ dataId: e, shape: r, dtype: i }], i), f = this.readToGPU(h, t);
      return this.disposeIntermediateTensorInfo(h), f;
    }
    if (u == null)
      throw s != null ? new Error("Data is not on GPU but on CPU.") : new Error("There is no data on GPU or CPU.");
    let l = this.decode(e, t.customTexShape), c = ds().makeTensorFromTensorInfo(l), p = this.texData.get(l.dataId);
    return { tensorRef: c, ...p.texture };
  }
  bufferSync(e) {
    let t = this.readSync(e.dataId);
    if (e.dtype === "string")
      try {
        let n = t.map((s) => w.decodeString(s));
        return Ae(e.shape, e.dtype, n);
      } catch (n) {
        throw new Error("Failed to decode encoded string bytes into utf-8");
      }
    return Ae(e.shape, e.dtype, t);
  }
  checkNumericalProblems(e) {
    if (e != null)
      for (let t = 0; t < e.length; t++) {
        let n = e[t];
        if (!QC(n))
          throw K().getBool("WEBGL_RENDER_FLOAT32_CAPABLE") ? Error(`The value ${n} cannot be represented with your current settings. Consider enabling float32 rendering: 'tf.env().set('WEBGL_RENDER_FLOAT32_ENABLED', true);'`) : Error(`The value ${n} cannot be represented on this device.`);
      }
  }
  getValuesFromTexture(e) {
    let { shape: t, dtype: n, isPacked: s } = this.texData.get(e), r = w.sizeFromShape(t);
    if (K().getBool("WEBGL_DOWNLOAD_FLOAT_ENABLED")) {
      let p = this.decode(e), d = this.texData.get(p.dataId), h = this.gpgpu.downloadMatrixFromPackedTexture(d.texture.texture, ...Jc(t)).subarray(0, r);
      return this.disposeIntermediateTensorInfo(p), h;
    }
    let a = K().getBool("WEBGL_PACK") && s === true, i = a ? ld(t) : t, o = a ? new QK(i) : new YK(i), u = this.runWebGLProgram(o, [{ shape: i, dtype: n, dataId: e }], "float32"), l = this.texData.get(u.dataId), c = this.gpgpu.downloadByteEncodedFloatMatrixFromOutputTexture(l.texture.texture, l.texShape[0], l.texShape[1]).subarray(0, r);
    return this.disposeIntermediateTensorInfo(u), c;
  }
  timerAvailable() {
    return K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE") > 0;
  }
  time(e) {
    let t = this.activeTimers, n = [], s = false;
    this.programTimersStack == null ? (this.programTimersStack = n, s = true) : this.activeTimers.push(n), this.activeTimers = n, e();
    let r = w.flatten(this.activeTimers.map((o) => o.query)).filter((o) => o != null), a = w.flatten(this.activeTimers.map((o) => o.name)).filter((o) => o != null);
    this.activeTimers = t, s && (this.programTimersStack = null);
    let i = { uploadWaitMs: this.uploadWaitMs, downloadWaitMs: this.downloadWaitMs, kernelMs: null, wallMs: null };
    return (async () => {
      if (K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE") > 0) {
        let o = await Promise.all(r);
        i.kernelMs = w.sum(o), i.getExtraProfileInfo = () => o.map((u, l) => ({ name: a[l], ms: u })).map((u) => `${u.name}: ${u.ms}`).join(", ");
      } else
        i.kernelMs = { error: "WebGL query timers are not supported in this environment." };
      return this.uploadWaitMs = 0, this.downloadWaitMs = 0, i;
    })();
  }
  memory() {
    return { unreliable: false, numBytesInGPU: this.numBytesInGPU, numBytesInGPUAllocated: this.textureManager.numBytesAllocated, numBytesInGPUFree: this.textureManager.numBytesFree };
  }
  startTimer() {
    return K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE") > 0 ? this.gpgpu.beginQuery() : { startMs: w.now(), endMs: null };
  }
  endTimer(e) {
    return K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE") > 0 ? (this.gpgpu.endQuery(), e) : (e.endMs = w.now(), e);
  }
  async getQueryTime(e) {
    if (K().getNumber("WEBGL_DISJOINT_QUERY_TIMER_EXTENSION_RELIABLE") > 0)
      return this.gpgpu.waitForQueryAndGetTime(e);
    let t = e;
    return t.endMs - t.startMs;
  }
  disposeData(e, t = false) {
    if (this.pendingDisposal.has(e))
      return false;
    if (!this.texData.has(e))
      return true;
    if (t ? this.texData.get(e).refCount = 0 : this.texData.get(e).refCount--, !t && this.texData.get(e).refCount > 0)
      return false;
    if (this.pendingRead.has(e))
      return this.pendingDisposal.add(e), this.pendingDeletes++, false;
    this.releaseGPUData(e);
    let { complexTensorInfos: n } = this.texData.get(e);
    return n != null && (this.disposeData(n.real.dataId, t), this.disposeData(n.imag.dataId, t)), this.texData.delete(e), true;
  }
  releaseGPUData(e) {
    let { texture: t, dtype: n, texShape: s, usage: r, isPacked: a, slice: i } = this.texData.get(e), o = i && i.origDataId || e, u = this.dataRefCount.get(o);
    u > 1 ? this.dataRefCount.set(o, u - 1) : (this.dataRefCount.delete(o), t != null && (this.numBytesInGPU -= this.computeBytes(s, n), this.textureManager.releaseTexture(t, s, r, a)));
    let l = this.texData.get(e);
    l.texture = null, l.texShape = null, l.isPacked = false, l.slice = null;
  }
  getTexture(e) {
    return this.uploadToGPU(e), this.texData.get(e).texture.texture;
  }
  getDataInfo(e) {
    return this.texData.get(e);
  }
  shouldExecuteOnCPU(e, t = l8) {
    return K().getBool("WEBGL_CPU_FORWARD") && e.every((n) => this.texData.get(n.dataId).texture == null && w.sizeFromShape(n.shape) < t);
  }
  getGPGPUContext() {
    return this.gpgpu;
  }
  where(e) {
    C.warn("tf.where() in webgl locks the UI thread. Call tf.whereAsync() instead");
    let t = e.dataSync();
    return a8(e.shape, t);
  }
  packedUnaryOp(e, t, n) {
    let s = new Jr(e.shape, t), r = this.compileAndRun(s, [e], n);
    return ds().makeTensorFromTensorInfo(r);
  }
  abs(e) {
    if (this.shouldExecuteOnCPU([e]) && e.dtype !== "complex64") {
      let s = U1(this.texData.get(e.dataId).values);
      return this.makeOutput(e.shape, e.dtype, s);
    }
    if (K().getBool("WEBGL_PACK_UNARY_OPERATIONS"))
      return this.packedUnaryOp(e, fw, e.dtype);
    let t = new Gs(e.shape, fw), n = this.compileAndRun(t, [e]);
    return ds().makeTensorFromTensorInfo(n);
  }
  makeTensorInfo(e, t, n) {
    let s;
    if (t === "string" && n != null && n.length > 0 && w.isString(n[0])) {
      let r = n.map((a) => w.encodeString(a));
      s = this.write(r, e, t);
    } else
      s = this.write(n, e, t);
    return this.texData.get(s).usage = null, { dataId: s, shape: e, dtype: t };
  }
  makeOutput(e, t, n) {
    return ds().makeTensorFromTensorInfo(this.makeTensorInfo(e, t, n), this);
  }
  unpackTensor(e) {
    let t = new r8(e.shape);
    return this.runWebGLProgram(t, [e], e.dtype);
  }
  packTensor(e) {
    let t = new WX(e.shape), n = true;
    return this.runWebGLProgram(t, [e], e.dtype, null, n);
  }
  packedReshape(e, t) {
    let n = [ya(e.shape), ...va(e.shape)], s = { dtype: e.dtype, shape: n, dataId: e.dataId }, r = [ya(t), ...va(t)], a = new q1(r, n), i = true, o = [n], u = this.runWebGLProgram(a, [s], e.dtype, o, i);
    return { dataId: u.dataId, shape: t, dtype: u.dtype };
  }
  decode(e, t) {
    let n = this.texData.get(e), { isPacked: s, shape: r, dtype: a } = n;
    if (t != null) {
      let p = w.sizeFromShape(r), d = t[0] * t[1] * 4;
      w.assert(p <= d, () => "customTexShape is too small. Row * Column * 4 should be equal or larger than the size of the tensor data.");
    }
    let i = ld(r), o;
    s ? o = new XK(i) : o = new KK(i);
    let u = true, l = [t != null ? t : Jc(i)], c = this.runWebGLProgram(o, [{ shape: i, dtype: a, dataId: e }], a, l, u, t);
    return { dtype: a, shape: r, dataId: c.dataId };
  }
  runWebGLProgram(e, t, n, s, r = false, a) {
    let i = this.makeTensorInfo(e.outputShape, n), o = this.texData.get(i.dataId);
    if (e.packedOutput && (o.isPacked = true), e.outPackingScheme === 0) {
      let g = a != null ? a : Jc(e.outputShape);
      o.texShape = g.map((b) => b * 2);
    }
    if (e.outTexUsage != null && (o.usage = e.outTexUsage), w.sizeFromShape(i.shape) === 0)
      return o.values = w.getTypedArrayFromDType(i.dtype, 0), i;
    let u = [], l = t.map((g) => {
      if (g.dtype === "complex64")
        throw new Error("GPGPUProgram does not support complex64 input. For complex64 dtypes, please separate the program into real and imaginary parts.");
      let b = this.texData.get(g.dataId);
      if (b.texture == null) {
        if (!e.packedInputs && w.sizeFromShape(g.shape) <= K().getNumber("WEBGL_SIZE_UPLOAD_UNIFORM"))
          return { shape: g.shape, texData: null, isUniform: true, uniformValues: b.values };
        e.packedInputs && (b.isPacked = true, b.shape = g.shape);
      }
      if (this.uploadToGPU(g.dataId), !!b.isPacked != !!e.packedInputs)
        g = b.isPacked ? this.unpackTensor(g) : this.packTensor(g), u.push(g), b = this.texData.get(g.dataId);
      else if (b.isPacked && !rl(b.shape, g.shape)) {
        let y = g, v = g.shape;
        g.shape = b.shape, g = this.packedReshape(g, v), u.push(g), b = this.texData.get(g.dataId), y.shape = v;
      }
      return { shape: g.shape, texData: b, isUniform: false };
    });
    this.uploadToGPU(i.dataId);
    let c = { shape: i.shape, texData: o, isUniform: false }, p = jK(e, l, c), d = this.getAndSaveBinary(p, () => HK(this.gpgpu, e, l, c)), h = this.activeTimers != null, f;
    h && (f = this.startTimer()), K().get("ENGINE_COMPILE_ONLY") || qK(this.gpgpu, d, l, c, s), u.forEach((g) => this.disposeIntermediateTensorInfo(g)), h && (f = this.endTimer(f), this.activeTimers.push({ name: e.constructor.name, query: this.getQueryTime(f) }));
    let m = K().get("WEBGL_FLUSH_THRESHOLD");
    if (m > 0) {
      let g = w.now();
      g - this.lastGlFlushTime > m && (this.gpgpu.gl.flush(), this.lastGlFlushTime = g);
    }
    if (!K().getBool("WEBGL_LAZILY_UNPACK") && o.isPacked && r === false) {
      let g = this.unpackTensor(i);
      return this.disposeIntermediateTensorInfo(i), g;
    }
    return i;
  }
  compileAndRun(e, t, n, s, r = false) {
    return n = n || t[0].dtype, this.runWebGLProgram(e, t, n, s, r);
  }
  getAndSaveBinary(e, t) {
    return e in this.binaryCache || (this.binaryCache[e] = t()), this.binaryCache[e];
  }
  getTextureManager() {
    return this.textureManager;
  }
  dispose() {
    this.disposed || (K().getBool("IS_TEST") || Object.keys(this.binaryCache).forEach((t) => {
      this.gpgpu.deleteProgram(this.binaryCache[t].webGLProgram), delete this.binaryCache[t];
    }), this.textureManager.dispose(), this.canvas != null && typeof HTMLCanvasElement != "undefined" && this.canvas instanceof HTMLCanvasElement ? this.canvas.remove() : this.canvas = null, this.gpgpuCreatedLocally && (this.gpgpu.program = null, this.gpgpu.dispose()), this.disposed = true);
  }
  floatPrecision() {
    return this.floatPrecisionValue == null && (this.floatPrecisionValue = j(() => {
      if (!K().get("WEBGL_RENDER_FLOAT32_ENABLED")) {
        let e = K().getBool("DEBUG");
        K().set("DEBUG", false);
        let t = this.abs(we(1e-8)).dataSync()[0];
        if (K().set("DEBUG", e), t > 0)
          return 32;
      }
      return 16;
    })), this.floatPrecisionValue;
  }
  epsilon() {
    return this.floatPrecision() === 32 ? i8 : o8;
  }
  uploadToGPU(e) {
    let t = this.texData.get(e), { shape: n, dtype: s, values: r, texture: a, usage: i, isPacked: o } = t;
    if (a != null)
      return;
    let u = this.activeTimers != null, l;
    u && (l = w.now());
    let c = t.texShape;
    if (c == null && (c = f1(n, o), t.texShape = c), r != null) {
      let p = ld(n), d, h = c[1], f = c[0], m = r instanceof Uint8Array || r instanceof Uint8ClampedArray;
      (o || !m) && ([h, f] = au(c[0], c[1])), o ? d = new JK(p, m) : d = new ZK(p, m);
      let g = m ? [f, h] : c, b = this.makeTensorInfo(g, s), y = this.texData.get(b.dataId);
      m ? y.usage = 2 : y.usage = 1, y.texShape = g, this.gpgpu.uploadDenseMatrixToTexture(this.getTexture(b.dataId), h, f, r);
      let v = [[f, h]], x = true, k = this.runWebGLProgram(d, [b], s, v, x), I = this.texData.get(k.dataId);
      t.texShape = I.texShape, t.isPacked = I.isPacked, t.usage = I.usage, K().get("ENGINE_COMPILE_ONLY") ? this.disposeData(k.dataId) : (t.texture = I.texture, t.values = null, this.texData.delete(k.dataId)), this.disposeIntermediateTensorInfo(b), u && (this.uploadWaitMs += w.now() - l);
    } else {
      let p = this.acquireTexture(c, i, s, o);
      t.texture = p;
    }
  }
  convertAndCacheOnCPU(e, t) {
    let n = this.texData.get(e), { dtype: s } = n;
    return this.releaseGPUData(e), t != null && (n.values = p8(t, s)), n.values;
  }
  acquireTexture(e, t, n, s) {
    if (this.numBytesInGPU += this.computeBytes(e, n), !this.warnedAboutMemory && this.numBytesInGPU > this.numMBBeforeWarning * 1024 * 1024) {
      let r = (this.numBytesInGPU / 1024 / 1024).toFixed(2);
      this.warnedAboutMemory = true, console.warn(`High memory usage in GPU: ${r} MB, most likely due to a memory leak`);
    }
    return this.textureManager.acquireTexture(e, t, s);
  }
  computeBytes(e, t) {
    return e[0] * e[1] * w.bytesPerElement(t);
  }
  checkCompileCompletion() {
    for (let [, e] of Object.entries(this.binaryCache))
      this.checkCompletion_(e);
  }
  async checkCompileCompletionAsync() {
    let e = [];
    if (this.gpgpu.parallelCompilationExtension) {
      for (let [, t] of Object.entries(this.binaryCache))
        e.push(this.checkCompletionAsync_(t));
      return Promise.all(e);
    } else {
      for (let [, t] of Object.entries(this.binaryCache)) {
        let n = new Promise((s) => {
          try {
            this.checkCompletion_(t), s(true);
          } catch (r) {
            throw r;
          }
        });
        e.push(n);
      }
      return Promise.all(e);
    }
  }
  async checkCompletionAsync_(e) {
    return this.gpgpu.gl.getProgramParameter(e.webGLProgram, this.gpgpu.parallelCompilationExtension.COMPLETION_STATUS_KHR) ? this.checkCompletion_(e) : (await HS(), this.checkCompletionAsync_(e));
  }
  checkCompletion_(e) {
    if (this.gpgpu.gl.getProgramParameter(e.webGLProgram, this.gpgpu.gl.LINK_STATUS) === false)
      throw console.log(this.gpgpu.gl.getProgramInfoLog(e.webGLProgram)), this.gpgpu.gl.getShaderParameter(e.fragmentShader, this.gpgpu.gl.COMPILE_STATUS) === false ? (vv(e.source, this.gpgpu.gl.getShaderInfoLog(e.fragmentShader)), new Error("Failed to compile fragment shader.")) : new Error("Failed to link vertex and fragment shaders.");
    return true;
  }
  getUniformLocations() {
    for (let [, e] of Object.entries(this.binaryCache)) {
      let { uniformLocations: t, customUniformLocations: n, infLoc: s, nanLoc: r, inShapesLocations: a, inTexShapesLocations: i, outShapeLocation: o, outShapeStridesLocation: u, outTexShapeLocation: l } = C1(this.gpgpu, e.program, e.webGLProgram);
      e.uniformLocations = t, e.customUniformLocations = n, e.infLoc = s, e.nanLoc = r, e.inShapesLocations = a, e.inTexShapesLocations = i, e.outShapeLocation = o, e.outShapeStridesLocation = u, e.outTexShapeLocation = l;
    }
  }
};
var K1 = j1;
K1.nextDataId = 0;
function p8(e, t) {
  if (t === "float32" || t === "complex64")
    return e;
  if (t === "int32" || t === "bool") {
    let n = t === "int32" ? new Int32Array(e.length) : new Uint8Array(e.length);
    for (let s = 0; s < n.length; ++s)
      n[s] = Math.round(e[s]);
    return n;
  } else
    throw new Error(`Unknown dtype ${t}`);
}
var fhe = "0.0.0";
function h8() {
  K().set("WEBGL_FORCE_F16_TEXTURES", true);
}
yp.isBrowser() && vp("webgl", () => new K1(), 2);
var mhe = { forceHalfFloat: h8 };
var X1 = `
  if (isnan(a)) return a;
  if (isnan(b)) return b;
`;
var lo = class {
  constructor(e, t, n) {
    this.variableNames = ["A", "B"], this.outputShape = C.assertAndGetBroadcastShape(t, n), this.enableShapeUniforms = Sn(this.outputShape.length), this.userCode = `
      float binaryOperation(float a, float b) {
        ${e}
      }

      void main() {
        float a = getAAtOutCoords();
        float b = getBAtOutCoords();
        setOutput(binaryOperation(a, b));
      }
    `;
  }
};
var eh = `
  result.r = isNaN.r > 0. ? NAN : result.r;
  result.g = isNaN.g > 0. ? NAN : result.g;
  result.b = isNaN.b > 0. ? NAN : result.b;
  result.a = isNaN.a > 0. ? NAN : result.a;
`;
var ec = class {
  constructor(e, t, n, s = false) {
    this.variableNames = ["A", "B"], this.supportsBroadcasting = true, this.packedInputs = true, this.packedOutput = true, this.outputShape = C.assertAndGetBroadcastShape(t, n);
    let r = this.outputShape.length;
    this.enableShapeUniforms = Sn(r);
    let a = "";
    if (s)
      if (r === 0 || w.sizeFromShape(this.outputShape) === 1)
        a = `
          result.y = 0.;
          result.z = 0.;
          result.w = 0.;
        `;
      else if (a = `
          ${ot(r)} coords = getOutputCoords();
        `, r === 1)
        this.enableShapeUniforms ? a += `
            result.y = (coords + 1) >= outShape ? 0. : result.y;
            result.z = 0.;
            result.w = 0.;
          ` : a += `
            result.y = (coords + 1) >= ${this.outputShape[0]} ? 0. : result.y;
            result.z = 0.;
            result.w = 0.;
          `;
      else {
        let o = ln("coords", r);
        this.enableShapeUniforms ? a += `
            bool nextRowOutOfBounds =
              (${o[r - 2]} + 1) >= outShape[${r} - 2];
            bool nextColOutOfBounds =
              (${o[r - 1]} + 1) >= outShape[${r} - 1];
            result.y = nextColOutOfBounds ? 0. : result.y;
            result.z = nextRowOutOfBounds ? 0. : result.z;
            result.w = nextColOutOfBounds || nextRowOutOfBounds ? 0. : result.w;
          ` : a += `
            bool nextRowOutOfBounds =
              (${o[r - 2]} + 1) >= ${this.outputShape[r - 2]};
            bool nextColOutOfBounds =
              (${o[r - 1]} + 1) >= ${this.outputShape[r - 1]};
            result.y = nextColOutOfBounds ? 0. : result.y;
            result.z = nextRowOutOfBounds ? 0. : result.z;
            result.w = nextColOutOfBounds || nextRowOutOfBounds ? 0. : result.w;
          `;
      }
    this.userCode = `
      vec4 binaryOperation(vec4 a, vec4 b) {
        ${e}
      }

      void main() {
        vec4 a = getAAtOutCoords();
        vec4 b = getBAtOutCoords();

        vec4 result = binaryOperation(a, b);
        ${a}

        setOutput(result);
      }
    `;
  }
};
function Rn(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  return n.incRef(s.dataId), { dataId: s.dataId, shape: s.shape, dtype: s.dtype };
}
var f8 = { kernelName: Wa, backendName: "webgl", kernelFunc: Rn };
function Rr(e) {
  let { inputs: t, backend: n } = e, { real: s, imag: r } = t, a = n.makeTensorInfo(s.shape, "complex64"), i = n.texData.get(a.dataId), o = Rn({ inputs: { x: s }, backend: n }), u = Rn({ inputs: { x: r }, backend: n });
  return i.complexTensorInfos = { real: o, imag: u }, a;
}
var m8 = { kernelName: ep, backendName: "webgl", kernelFunc: Rr };
var Y1 = "return (a < 0.) ? b * a : a;";
var Q1 = `
  vec4 aLessThanZero = vec4(lessThan(a, vec4(0.)));
  return (aLessThanZero * (b * a)) + ((vec4(1.0) - aLessThanZero) * a);
`;
function g8(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { alpha: a } = s, i = n.makeTensorInfo([], "float32", w.createScalarValue(a, "float32")), o = K().getBool("WEBGL_PACK_BINARY_OPERATIONS") ? new ec(Q1, r.shape, i.shape) : new lo(Y1, r.shape, i.shape), u = n.runWebGLProgram(o, [r, i], "float32");
  return n.disposeIntermediateTensorInfo(i), u;
}
var b8 = { kernelName: Ua, backendName: "webgl", kernelFunc: g8 };
var Z1 = "return (a < 0.) ? b * a : a;";
var J1 = `
  vec4 aLessThanZero = vec4(lessThan(a, vec4(0.)));
  return (aLessThanZero * (b * a)) + ((vec4(1.0) - aLessThanZero) * a);
`;
function y8(e) {
  let { inputs: t, backend: n } = e, { x: s, alpha: r } = t, a = K().getBool("WEBGL_PACK_BINARY_OPERATIONS") ? new ec(J1, s.shape, r.shape) : new lo(Z1, s.shape, r.shape);
  return n.runWebGLProgram(a, [s, r], "float32");
}
var v8 = { kernelName: ti, backendName: "webgl", kernelFunc: y8 };
var du = "if (isnan(x)) return x;";
var x8 = `
  if (isnan(a)) return a;
  if (isnan(b)) return b;
`;
var w8 = `
  result.r = isNaN.r > 0. ? NAN : result.r;
  result.g = isNaN.g > 0. ? NAN : result.g;
  result.b = isNaN.b > 0. ? NAN : result.b;
  result.a = isNaN.a > 0. ? NAN : result.a;
`;
function Ke({ opSnippet: e, packedOpSnippet: t, cpuKernelImpl: n, dtype: s }) {
  return ({ inputs: r, backend: a }) => {
    let { x: i } = r, o = a, u = s || i.dtype;
    if (o.shouldExecuteOnCPU([i]) && n != null) {
      let p = o.texData.get(i.dataId), d = n(p.values, u);
      return o.makeTensorInfo(i.shape, u, d);
    }
    let l = K().getBool("WEBGL_PACK_UNARY_OPERATIONS") && t != null, c;
    return l ? c = new Jr(i.shape, t) : c = new Gs(i.shape, e), o.runWebGLProgram(c, [i], u);
  };
}
function jt({ opSnippet: e, packedOpSnippet: t, checkOutOfBounds: n = false, supportsComplex: s = false, cpuKernelImpl: r, dtype: a }) {
  return ({ inputs: i, backend: o }) => {
    let { a: u, b: l } = i, c = o;
    if (s && u.dtype === "complex64") {
      let f = c.texData.get(u.dataId), m = c.texData.get(l.dataId), [g, b] = [[f.complexTensorInfos.real, m.complexTensorInfos.real], [f.complexTensorInfos.imag, m.complexTensorInfos.imag]].map((v) => {
        let [x, k] = v, I = { dataId: x.dataId, dtype: x.dtype, shape: u.shape }, $ = { dataId: k.dataId, dtype: k.dtype, shape: l.shape }, R = new lo(e, u.shape, l.shape);
        return c.runWebGLProgram(R, [I, $], cn(x.dtype, k.dtype));
      }), y = Rr({ inputs: { real: g, imag: b }, backend: c });
      return c.disposeIntermediateTensorInfo(g), c.disposeIntermediateTensorInfo(b), y;
    }
    let p = a || cn(u.dtype, l.dtype);
    if ((u.dtype === "string" || l.dtype === "string" || c.shouldExecuteOnCPU([u, l])) && r != null) {
      let f = c.texData.get(u.dataId).values, m = c.texData.get(l.dataId).values, g = u.dtype === "string" ? C.fromUint8ToStringArray(f) : f, b = u.dtype === "string" ? C.fromUint8ToStringArray(m) : m, [y, v] = r(u.shape, l.shape, g, b, p), x = c.makeTensorInfo(v, p), k = c.texData.get(x.dataId);
      return k.values = y, x;
    }
    let d = K().getBool("WEBGL_PACK_BINARY_OPERATIONS") && t != null, h;
    return d ? h = new ec(t, u.shape, l.shape, n) : h = new lo(e, u.shape, l.shape), c.runWebGLProgram(h, [u, l], p);
  };
}
function th(e, t = false) {
  if (e === "linear")
    return t ? JX : KX;
  if (e === "relu")
    return t ? t8 : YX;
  if (e === "elu")
    return t ? e8 : XX;
  if (e === "relu6")
    return t ? n8 : QX;
  if (e === "prelu")
    return t ? J1 : Z1;
  if (e === "leakyrelu")
    return t ? Q1 : Y1;
  if (e === "sigmoid")
    return t ? s8 : ZX;
  throw new Error(`Activation ${e} has not been implemented for the WebGL backend.`);
}
var e2 = class {
  constructor(e, t, n, s = false, r = false, a = false, i = null, o = false, u = false) {
    this.variableNames = ["matrixA", "matrixB"], this.packedInputs = true, this.packedOutput = true, this.outputShape = n, this.enableShapeUniforms = Sn(this.outputShape.length);
    let l = s ? e[1] : e[2], c = Math.ceil(l / 2), p = s ? "i * 2, rc.y" : "rc.y, i * 2", d = r ? "rc.z, i * 2" : "i * 2, rc.z", h = s ? ["a.xxyy", "a.zzww"] : ["a.xxzz", "a.yyww"], f = r ? ["b.xzxz", "b.ywyw"] : ["b.xyxy", "b.zwzw"], m = "", g = "";
    i && (o ? m = `vec4 activation(vec4 a) {
          vec4 b = getPreluActivationWeightsAtOutCoords();
          ${i}
        }` : u ? m = `vec4 activation(vec4 a) {
          vec4 b = getLeakyreluAlphaAtOutCoords();
          ${i}
        }` : m = `vec4 activation(vec4 x) {
          ${i}
        }`, g = "result = activation(result);");
    let b = a ? "result += getBiasAtOutCoords();" : "";
    a && this.variableNames.push("bias"), o && this.variableNames.push("preluActivationWeights"), u && this.variableNames.push("leakyreluAlpha");
    let y = "rc.x", v = "rc.x";
    e[0] < t[0] ? y = `int(min(float(rc.x), ${e[0] - 1}.))` : t[0] < e[0] && (v = `int(min(float(rc.x), ${t[0] - 1}.))`), this.userCode = `
      ${m}
      // Don't use uniform for sharedDimensionPacked for performance.
      const float sharedDimension = ${c}.0;

      vec4 dot2x2ARowBCol(ivec3 rc) {
        vec4 result = vec4(0);
        for (int i = 0; i < ${c}; i++) {
          int batchA = ${y};
          int batchB = ${v};
          vec4 a = getMatrixA(batchA, ${p});
          vec4 b = getMatrixB(batchB, ${d});

          // These swizzled products need to be separately added.
          // See: https://github.com/tensorflow/tfjs/issues/1735
          result += (${h[0]} * ${f[0]});
          result += (${h[1]} * ${f[1]});
        }
        return result;
      }

      void main() {
        ivec3 rc = getOutputCoords();
        vec4 result = dot2x2ARowBCol(rc);

        ${b}

        ${g}

        setOutput(result);
      }
    `;
  }
};
var mw = { REAL: "return areal * breal - aimag * bimag;", IMAG: "return areal * bimag + aimag * breal;" };
var gw = class {
  constructor(e, t, n) {
    this.variableNames = ["AReal", "AImag", "BReal", "BImag"], this.outputShape = C.assertAndGetBroadcastShape(t, n), this.userCode = `
      float binaryOpComplex(
          float areal, float aimag, float breal, float bimag) {
        ${e}
      }

      void main() {
        float areal = getARealAtOutCoords();
        float aimag = getAImagAtOutCoords();
        float breal = getBRealAtOutCoords();
        float bimag = getBImagAtOutCoords();
        setOutput(binaryOpComplex(areal, aimag, breal, bimag));
      }
    `;
  }
};
var bw = "return a * b;";
function _v(e) {
  let { inputs: t, backend: n } = e, { a: s, b: r } = t, a = C.upcastType(s.dtype, r.dtype);
  if (s.dtype === "complex64") {
    let o = n.texData.get(s.dataId), u = n.texData.get(r.dataId), l = new gw(mw.REAL, s.shape, r.shape), c = new gw(mw.IMAG, s.shape, r.shape), p = [{ dataId: o.complexTensorInfos.real.dataId, dtype: o.complexTensorInfos.real.dtype, shape: s.shape }, { dataId: o.complexTensorInfos.imag.dataId, dtype: o.complexTensorInfos.imag.dtype, shape: s.shape }, { dataId: u.complexTensorInfos.real.dataId, dtype: u.complexTensorInfos.real.dtype, shape: r.shape }, { dataId: u.complexTensorInfos.imag.dataId, dtype: u.complexTensorInfos.imag.dtype, shape: r.shape }], d = n.runWebGLProgram(l, p, "float32"), h = n.runWebGLProgram(c, p, "float32"), f = Rr({ inputs: { real: d, imag: h }, backend: n });
    return n.disposeIntermediateTensorInfo(d), n.disposeIntermediateTensorInfo(h), f;
  }
  if (n.shouldExecuteOnCPU([s, r])) {
    let o = n.texData.get(s.dataId), u = n.texData.get(r.dataId), [l, c] = wX(s.shape, r.shape, o.values, u.values, a), p = n.makeTensorInfo(c, a), d = n.texData.get(p.dataId);
    return d.values = l, p;
  }
  let i;
  return K().getBool("WEBGL_PACK_BINARY_OPERATIONS") ? i = new ec(bw, s.shape, r.shape) : i = new lo(bw, s.shape, r.shape), n.runWebGLProgram(i, [s, r], a);
}
var k8 = { kernelName: Za, backendName: "webgl", kernelFunc: _v };
function S8(e, t, n) {
  let s = [ya(e.shape), ...va(e.shape)], r = { dtype: e.dtype, shape: s, dataId: e.dataId }, a = [ya(t), ...va(t)], i = new q1(a, s), o = true, u = [s], l = n.runWebGLProgram(i, [r], e.dtype, u, o);
  return { dataId: l.dataId, shape: t, dtype: l.dtype };
}
function he(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { shape: a } = s, i = n, o = w.sizeFromShape(r.shape), u = w.inferFromImplicitShape(a, o), l = w.sizeFromShape(u);
  w.assert(o === l, () => `The new shape (${u}) has ${l} elements and the old shape (${r.shape}) has ${o} elements. The new shape and old shape must have the same number of elements.`);
  let c = i.texData.get(r.dataId);
  return c.isPacked && !rl(r.shape, u) && !(c.texture !== null && rl(c.shape, u)) ? S8(r, u, i) : (i.incRef(r.dataId), { dataId: r.dataId, shape: u, dtype: r.dtype });
}
var I8 = { kernelName: Oo, backendName: "webgl", kernelFunc: he };
var yw = class {
  constructor(e, t) {
    this.variableNames = ["x"];
    let { windowSize: n, batchSize: s, inSize: r, outSize: a } = e;
    this.outputShape = [s, a];
    let i = Math.floor(n / 4) * 4, o = n % 4, u = "sumValue += dot(values, ones);";
    if (t != null) {
      let c = 1 / t;
      u = `sumValue += dot(values * ${w.isInt(c) ? c.toPrecision(2) : c}, ones);`;
    }
    let l = "";
    r % n > 0 && (l = `
        if (inIdx < 0 || inIdx >= ${r}) {
          return 0.0;
        }
      `), this.userCode = `
      const vec4 ones = vec4(1.0, 1.0, 1.0, 1.0);

      float getValue(int batch, int inIdx) {
        ${l}
        return getX(batch, inIdx);
      }

      void main() {
        ivec2 coords = getOutputCoords();
        int batch = coords[0];
        int outIdx = coords[1];
        int inOffset = outIdx * ${n};

        float sumValue = 0.0;

        for (int i = 0; i < ${i}; i += 4) {
          int inIdx = inOffset + i;
          vec4 values = vec4(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1),
            getValue(batch, inIdx + 2),
            getValue(batch, inIdx + 3)
          );

          ${u}
        }

        int inIdx = inOffset + ${i};
        if (${o === 1}) {
          vec4 values = vec4(getValue(batch, inIdx), 0.0, 0.0, 0.0);

          ${u}
        } else if (${o === 2}) {
          vec4 values = vec4(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1), 0.0, 0.0);

          ${u}
        } else if (${o === 3}) {
          vec4 values = vec4(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1),
            getValue(batch, inIdx + 2), 0.0);

          ${u}
        }
        setOutput(sumValue);
      }
    `;
  }
};
var C8 = class {
  constructor(e, t) {
    this.variableNames = ["x"];
    let { windowSize: n, batchSize: s, inSize: r, outSize: a } = e;
    this.outputShape = [s, a];
    let i = "0.0", o = "";
    t === "prod" ? i = "1.0" : t === "min" ? (i = "1.0 / 1e-20", o = "min") : t === "max" && (i = "-1.0 / 1e-20", o = "max");
    let u = `${t}(${t}(${t}(minMaxValue[0], minMaxValue[1]), minMaxValue[2]), minMaxValue[3])`;
    t === "sum" ? u = "sumValue" : t === "prod" ? u = "prodValue" : t === "all" ? u = "allValue" : t === "any" && (u = "anyValue");
    let l = Math.floor(n / 4) * 4, c = n % 4, p = `
      if (${t === "sum"}) {
        sumValue += dot(values, ones);
      } else if (${t === "prod"}) {
        vec2 tmp = vec2(values[0], values[1]) * vec2(values[2], values[3]);
        prodValue *= tmp[0] * tmp[1];
      } else {
        minMaxValue = ${o}(values, minMaxValue);
        if (${t === "min"} || ${t === "max"}) {
          minMaxValue = ${o}(values, minMaxValue);
          bvec4 isNaN = isnan(values);
          if (isNaN.r || isNaN.g || isNaN.b || isNaN.a) {
            minMaxValue = vec4(NAN);
          }
        }
      }
    `, d = "vec4";
    t === "all" ? (i = "1.0", p = `
        bool reducedAllValue = all(values);
        float floatedReducedAllValue = float(reducedAllValue);
        allValue = float(allValue >= 1.0 && floatedReducedAllValue >= 1.0);
      `, d = "bvec4") : t === "any" && (i = "0.0", p = `
        bool reducedAnyValue = any(values);
        float floatedReducedAnyValue = float(reducedAnyValue);
        anyValue = float(anyValue >= 1.0 || floatedReducedAnyValue >= 1.0);
      `, d = "bvec4");
    let h = "";
    r % n > 0 && (h = `
        if (inIdx < 0 || inIdx >= ${r}) {
          return initializationValue;
        }
      `), this.userCode = `
      const float initializationValue = ${i};
      const vec4 ones = vec4(1.0, 1.0, 1.0, 1.0);

      float getValue(int batch, int inIdx) {
        ${h}
        return getX(batch, inIdx);
      }

      void main() {
        ivec2 coords = getOutputCoords();
        int batch = coords[0];
        int outIdx = coords[1];
        int inOffset = outIdx * ${n};

        vec4 minMaxValue = vec4(${i});
        float prodValue = 1.0;
        float sumValue = 0.0;
        float allValue = 1.0;
        float anyValue = 0.0;

        for (int i = 0; i < ${l}; i += 4) {
          int inIdx = inOffset + i;
          ${d} values = ${d}(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1),
            getValue(batch, inIdx + 2),
            getValue(batch, inIdx + 3)
          );

          ${p}
        }

        int inIdx = inOffset + ${l};
        if (${c === 1}) {
          ${d} values = ${d}(
            getValue(batch, inIdx),
            initializationValue,
            initializationValue,
            initializationValue
          );

          ${p}
        } else if (${c === 2}) {
          ${d} values = ${d}(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1),
            initializationValue,
            initializationValue
          );

          ${p}
        } else if (${c === 3}) {
          ${d} values = ${d}(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1),
            getValue(batch, inIdx + 2),
            initializationValue
          );

          ${p}
        }
        setOutput(${u});
      }
    `;
  }
};
function N8(e) {
  let t = [];
  for (; t.length === 0 || t[t.length - 1].outSize !== 1; ) {
    let n = t.length ? t[t.length - 1].outSize : e[1], s = C.computeOptimalWindowSize(n);
    t.push({ inSize: n, windowSize: s, outSize: Math.ceil(n / s) });
  }
  return t;
}
function Si(e, t, n, s) {
  let r = N8(e.shape), a = e;
  for (let i = 0; i < r.length; i++) {
    let { inSize: o, windowSize: u, outSize: l } = r[i], c, p;
    n === "mean" ? c = i === 0 ? new yw({ windowSize: u, inSize: o, batchSize: e.shape[0], outSize: l }, o) : new yw({ windowSize: u, inSize: o, batchSize: e.shape[0], outSize: l }) : c = new C8({ windowSize: u, inSize: o, batchSize: e.shape[0], outSize: l }, n), p = a, a = s.runWebGLProgram(c, [a], t), p.dataId !== e.dataId && s.disposeIntermediateTensorInfo(p);
  }
  return a;
}
var T8 = class {
  constructor(e, t) {
    this.variableNames = ["A"];
    let n = new Array(e.length);
    for (let a = 0; a < n.length; a++)
      n[a] = e[t[a]];
    this.outputShape = n, this.rank = n.length;
    let s = ot(this.rank), r = $8(t);
    this.userCode = `
    void main() {
      ${s} resRC = getOutputCoords();
      setOutput(getA(${r}));
    }
    `;
  }
};
function $8(e) {
  let t = e.length;
  if (t > 6)
    throw Error(`Transpose for rank ${t} is not yet supported`);
  let n = ["resRC.x", "resRC.y", "resRC.z", "resRC.w", "resRC.u", "resRC.v"], s = new Array(t);
  for (let r = 0; r < e.length; r++)
    s[e[r]] = n[r];
  return s.join();
}
var _8 = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true;
    let n = new Array(e.length);
    for (let l = 0; l < n.length; l++)
      n[l] = e[t[l]];
    if (this.outputShape = n, this.rank = n.length, this.rank > 6)
      throw Error(`Packed transpose for rank ${this.rank} is not yet supported.`);
    let s = ot(this.rank), r = H1("rc", this.rank), a = new Array(this.rank);
    for (let l = 0; l < t.length; l++)
      a[t[l]] = r[l];
    let i = `vec2(${a.slice(-2).join()})`, o = `++${r[this.rank - 1]} < ${n[this.rank - 1]}`, u = `getChannel(getA(${a.join()}), ${i})`;
    this.userCode = `
    void main() {
      ${s} rc = getOutputCoords();
      vec4 result = vec4(0.);
      result[0] = ${u};
      if(${o}) {
        result[1] = ${u};
      }
      --${r[this.rank - 1]};
      if(++${r[this.rank - 2]} < ${n[this.rank - 2]}) {
        result[2] = ${u};
        if(${o}) {
          result[3] = ${u};
        }
      }
      setOutput(result);
    }
    `;
  }
};
function nh(e, t, n) {
  let s = K().getBool("WEBGL_PACK_ARRAY_OPERATIONS") ? new _8(e.shape, t) : new T8(e.shape, t);
  return n.runWebGLProgram(s, [e], e.dtype);
}
function A8(e, t, n, s) {
  let r = t, a = e.shape.length, i = w.parseAxisParam(r, e.shape), o = i, u = C.getAxesPermutation(o, a), l = u != null, c = e;
  l && (c = nh(e, u, s), o = C.getInnerMostAxes(o.length, a)), C.assertAxesAreInnerMostDims("sum", o, a);
  let [p, d] = C.computeOutAndReduceShapes(c.shape, o), h = p;
  n && (h = C.expandShapeToKeepDim(p, i));
  let f = w.sizeFromShape(d), g = w.sizeFromShape(e.shape) / f, b = he({ inputs: { x: c }, attrs: { shape: [g, f] }, backend: s }), y = bp(e.dtype), v = Si(b, y, "sum", s), x = he({ inputs: { x: v }, attrs: { shape: h }, backend: s });
  return s.disposeIntermediateTensorInfo(b), s.disposeIntermediateTensorInfo(v), l && s.disposeIntermediateTensorInfo(c), x;
}
function sh(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  return A8(r, a, i, n);
}
var E8 = { kernelName: ci, backendName: "webgl", kernelFunc: sh };
function _t(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { perm: a } = s, i = n, o = r.shape.length, u = new Array(o);
  for (let c = 0; c < u.length; c++)
    u[c] = r.shape[a[c]];
  let l;
  if (i.shouldExecuteOnCPU([r])) {
    let p = i.texData.get(r.dataId).values, d = $v(p, r.shape, r.dtype, a, u);
    l = i.makeTensorInfo(u, r.dtype);
    let h = i.texData.get(l.dataId);
    h.values = d;
  } else
    l = nh(r, a, i);
  return l;
}
var R8 = { kernelName: mi, backendName: "webgl", kernelFunc: _t };
var t2 = 1e3;
function Gd({ a: e, b: t, transposeA: n, transposeB: s, backend: r, bias: a = null, preluActivationWeights: i = null, leakyreluAlpha: o = 0, activation: u = null }) {
  let l = e.shape.length, c = t.shape.length, p = n ? e.shape[l - 2] : e.shape[l - 1], d = s ? t.shape[c - 1] : t.shape[c - 2], h = n ? e.shape[l - 1] : e.shape[l - 2], f = s ? t.shape[c - 2] : t.shape[c - 1], m = e.shape.slice(0, -2), g = t.shape.slice(0, -2), b = w.sizeFromShape(m), y = w.sizeFromShape(g), x = Qo.assertAndGetBroadcastShape(e.shape.slice(0, -2), t.shape.slice(0, -2)).concat([h, f]);
  w.assert(p === d, () => `Error in matMul: inner shapes (${p}) and (${d}) of Tensors with shapes ${e.shape} and ${t.shape} and transposeA=${n} and transposeB=${s} must match.`);
  let k = n ? [b, p, h] : [b, h, p], I = s ? [y, f, d] : [y, d, f], $ = he({ inputs: { x: e }, backend: r, attrs: { shape: k } }), R = he({ inputs: { x: t }, backend: r, attrs: { shape: I } }), E = [$, R], P = Math.max(b, y), A = n ? $.shape[1] : $.shape[2], O = a != null, T = i != null, z = u === "leakyrelu", W = u != null ? th(u, true) : null, q = O || T || z || W != null, X;
  if ((h === 1 || f === 1) && A > t2 && q === false) {
    let Z = $, te = R;
    n && (Z = _t({ inputs: { x: $ }, backend: r, attrs: { perm: [0, 2, 1] } }), E.push(Z)), s && (te = _t({ inputs: { x: R }, backend: r, attrs: { perm: [0, 2, 1] } }), E.push(te));
    let J = f !== 1, se = f === 1, ne = Z;
    J && (ne = he({ inputs: { x: Z }, backend: r, attrs: { shape: [P, A, 1] } }), E.push(ne));
    let oe = f === 1 ? 2 : 1, ae = te;
    se && (ae = he({ inputs: { x: te }, backend: r, attrs: { shape: [P, 1, A] } }), E.push(ae));
    let de = _v({ inputs: { a: ne, b: ae }, backend: r });
    X = sh({ inputs: { x: de }, backend: r, attrs: { axis: oe, keepDims: true } }), E.push(de);
  } else {
    let Z = cn(e.dtype, t.dtype), te = new e2(k, I, [P, h, f], n, s, O, W, T, z), J = [$, R];
    if (a != null && J.push(a), T && J.push(i), z) {
      let se = r.makeTensorInfo([], "float32", w.createScalarValue(o, "float32"));
      J.push(se), E.push(se);
    }
    X = r.runWebGLProgram(te, J, Z);
  }
  let Y = he({ inputs: { x: X }, backend: r, attrs: { shape: x } });
  E.push(X);
  for (let Z of E)
    r.disposeIntermediateTensorInfo(Z);
  return Y;
}
function D8(e) {
  let { inputs: t, backend: n, attrs: s } = e, { a: r, b: a, bias: i, preluActivationWeights: o } = t, { transposeA: u, transposeB: l, activation: c, leakyreluAlpha: p } = s;
  return Gd({ a: r, b: a, transposeA: u, transposeB: l, backend: n, bias: i, preluActivationWeights: o, leakyreluAlpha: p, activation: c });
}
var F8 = { kernelName: aa, backendName: "webgl", kernelFunc: D8 };
var vw = "return abs(x);";
function O8(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (n.shouldExecuteOnCPU([s]) && s.dtype !== "complex64") {
    let a = n.texData.get(s.dataId), i = U1(a.values);
    return n.makeTensorInfo(s.shape, s.dtype, i);
  }
  let r;
  return K().getBool("WEBGL_PACK_UNARY_OPERATIONS") ? r = new Jr(s.shape, vw) : r = new Gs(s.shape, vw), n.runWebGLProgram(r, [s], s.dtype);
}
var P8 = { kernelName: po, backendName: "webgl", kernelFunc: O8 };
var z8 = ss + `
  if (abs(x) > 1.) {
    return NAN;
  }
  return acos(x);
`;
var M8 = Ke({ opSnippet: z8 });
var L8 = { kernelName: ol, backendName: "webgl", kernelFunc: M8 };
var B8 = ss + `
  if (x < 1.0) return NAN;
return log(x + sqrt(x * x - 1.0));`;
var V8 = Ke({ opSnippet: B8 });
var W8 = { kernelName: ul, backendName: "webgl", kernelFunc: V8 };
var xw = "return a + b;";
var U8 = jt({ opSnippet: xw, packedOpSnippet: xw, supportsComplex: true, cpuKernelImpl: nX });
var G8 = { kernelName: Sr, backendName: "webgl", kernelFunc: U8 };
var H8 = class {
  constructor(e, t) {
    this.outputShape = [], this.outputShape = e, this.variableNames = t.map((r, a) => `T${a}`);
    let n = [];
    this.variableNames.forEach((r) => {
      n.push(`float v${r} = get${r}AtOutCoords();`);
    });
    let s = this.variableNames.map((r) => `v${r}`).join(" + ");
    this.userCode = `
      void main() {
        ${n.join(`
        `)}

        float result = ${s};
        setOutput(result);
      }
    `;
  }
};
var q8 = class {
  constructor(e, t) {
    this.outputShape = [], this.packedInputs = true, this.packedOutput = true, this.outputShape = e, this.variableNames = t.map((r, a) => `T${a}`);
    let n = [];
    this.variableNames.forEach((r) => {
      n.push(`vec4 v${r} = get${r}AtOutCoords();`);
    });
    let s = this.variableNames.map((r) => `v${r}`).join(" + ");
    this.userCode = `
      void main() {
        ${n.join(`
        `)}

        vec4 result = ${s};
        setOutput(result);
      }
    `;
  }
};
function pd(e) {
  let { inputs: t, backend: n } = e, s = t;
  if (s.length === 1)
    return Rn({ inputs: { x: s[0] }, backend: n });
  if (s.length > K().get("WEBGL_MAX_TEXTURES_IN_SHADER")) {
    let u = Math.floor(s.length / 2), l = pd({ inputs: s.slice(0, u), backend: n }), c = pd({ inputs: s.slice(u), backend: n });
    return pd({ inputs: [l, c], backend: n });
  }
  let r = s.map((u) => u.dtype).reduce((u, l) => cn(u, l)), a = s.map((u) => u.shape), o = K().getBool("WEBGL_PACK") ? new q8(s[0].shape, a) : new H8(s[0].shape, a);
  return n.runWebGLProgram(o, s, r);
}
var j8 = { kernelName: Sa, backendName: "webgl", kernelFunc: pd };
function K8(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s, o = r.shape.length, u = w.parseAxisParam(a, r.shape), l = u, c = C.getAxesPermutation(l, o), p = r;
  c != null && (p = _t({ inputs: { x: r }, backend: n, attrs: { perm: c } }), l = C.getInnerMostAxes(l.length, o)), C.assertAxesAreInnerMostDims("all", l, o);
  let [d, h] = C.computeOutAndReduceShapes(p.shape, l), f = w.sizeFromShape(h), m = he({ inputs: { x: p }, backend: n, attrs: { shape: [-1, f] } }), g = Si(m, m.dtype, "all", n), b;
  if (i) {
    let y = C.expandShapeToKeepDim(d, u);
    b = he({ inputs: { x: g }, backend: n, attrs: { shape: y } });
  } else
    b = he({ inputs: { x: g }, backend: n, attrs: { shape: d } });
  return n.disposeIntermediateTensorInfo(m), n.disposeIntermediateTensorInfo(g), c != null && n.disposeIntermediateTensorInfo(p), b;
}
var X8 = { kernelName: ll, backendName: "webgl", kernelFunc: K8 };
function Y8(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s, o = r.shape.length, u = w.parseAxisParam(a, r.shape), l = u, c = C.getAxesPermutation(l, o), p = r;
  c != null && (p = _t({ inputs: { x: r }, backend: n, attrs: { perm: c } }), l = C.getInnerMostAxes(l.length, o)), C.assertAxesAreInnerMostDims("any", l, o);
  let [d, h] = C.computeOutAndReduceShapes(p.shape, l), f = w.sizeFromShape(h), m = he({ inputs: { x: p }, backend: n, attrs: { shape: [-1, f] } }), g = Si(m, m.dtype, "any", n), b;
  if (i) {
    let y = C.expandShapeToKeepDim(d, u);
    b = he({ inputs: { x: g }, backend: n, attrs: { shape: y } });
  } else
    b = he({ inputs: { x: g }, backend: n, attrs: { shape: d } });
  return n.disposeIntermediateTensorInfo(m), n.disposeIntermediateTensorInfo(g), c != null && n.disposeIntermediateTensorInfo(p), b;
}
var Q8 = { kernelName: cl, backendName: "webgl", kernelFunc: Y8 };
var Z8 = class {
  constructor(e, t, n) {
    this.variableNames = ["A"];
    let { windowSize: s, batchSize: r, outSize: a } = e;
    n || this.variableNames.push("bestIndicesA"), this.outputShape = [r, a];
    let i = t === "max" ? ">" : "<", o = n ? "inOffset + i;" : "round(getBestIndicesA(batch, inOffset + i));";
    this.userCode = `
      void main() {
        ivec2 coords = getOutputCoords();
        int batch = coords[0];
        int outIdx = coords[1];
        int inOffset = outIdx * ${s};

        int bestIndex = inOffset;
        float bestValue = getA(batch, bestIndex);

        for (int i = 0; i < ${s}; i++) {
          int inIdx = ${o};
          float candidate = getA(batch, inIdx);
          if (candidate ${i} bestValue) {
            bestValue = candidate;
            bestIndex = inIdx;
          }
        }
        setOutput(float(bestIndex));
      }
    `;
  }
};
var J8 = class {
  constructor(e, t, n, s) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true, w.assert(e.length > 2, () => `Packed arg${n.charAt(0).toUpperCase() + n.slice(1)} supports only inputs with rank above 2.`);
    let r = e[e.length - 1], a = Math.ceil(r / t);
    this.outputShape = e.slice(0, -1), a > 1 && this.outputShape.push(a), s || this.variableNames.push("bestIndicesA");
    let i = this.outputShape, o = i.length, u = ot(o), l = ln("coords", o), c, p;
    if (a === 1) {
      p = o + 1;
      let $ = ot(p);
      c = `
        ${$} sourceLocR = ${$}(${l.join()}, 0);
        ++${l[o - 1]};
        ${$} sourceLocG = ${$}(${l.join()}, 0);
        ++${l[o - 2]};
        ${$} sourceLocA = ${$}(${l.join()}, 0);
        --${l[o - 1]};
        ${$} sourceLocB = ${$}(${l.join()}, 0);
        --${l[o - 2]};`;
    } else
      p = o, c = `
        ${u} sourceLocR = coords;
        ++${l[o - 1]};
        ${u} sourceLocG = coords;
        ++${l[o - 2]};
        ${u} sourceLocA = coords;
        --${l[o - 1]};
        ${u} sourceLocB = coords;
        --${l[o - 2]};`;
    let d = ["x", "y", "z", "w", "u", "v"].slice(0, p), h = "." + d[p - 1], f = d.map(($) => "int " + $), m = ln("sourceLocR", p - 1).concat("inIdx.r"), g = ln("sourceLocG", p - 1).concat("inIdx.g"), b = ln("sourceLocB", p - 1).concat("inIdx.b"), y = ln("sourceLocA", p - 1).concat("inIdx.a"), v = n === "max" ? "greaterThan" : "lessThan", x = s ? "" : `
          inIdx = round(vec4(getBestIndicesAChannel(${m.join()}),
                             getBestIndicesAChannel(${g.join()}),
                             getBestIndicesAChannel(${b.join()}),
                             getBestIndicesAChannel(${y.join()})));`, k = `vec4(
            getAChannel(${m.join()}),
            hasNextCol ? getAChannel(${g.join()}) : 0.,
            hasNextRow ? getAChannel(${b.join()}) : 0.,
            hasNextRow && hasNextCol ? getAChannel(${y.join()}) : 0.)`, I = s ? "" : `
      float getBestIndicesAChannel(${f.join()}) {
        return getChannel(getBestIndicesA(${d.join()}),
                                          vec2(${d.slice(-2).join()}));
      }`;
    this.userCode = `
      float getAChannel(${f.join()}) {
        return getChannel(getA(${d.join()}),
                               vec2(${d.slice(-2).join()}));
      }
      ${I}
      void main() {
        ${u} coords = getOutputCoords();
        bool hasNextCol = ${l[o - 1]} < ${i[o - 1] - 1};
        bool hasNextRow = ${l[o - 2]} < ${i[o - 2] - 1};
        ${c}
        ivec4 srcIdx = ivec4(sourceLocR${h}, sourceLocG${h},
          sourceLocB${h}, sourceLocA${h}) * ${t};
        ivec4 inIdx = srcIdx;
        vec4 bestIndex = vec4(inIdx);
        vec4 bestValue = ${k};

        for (int i = 0; i < ${t}; i++) {
          inIdx = srcIdx;
          ${x}
          vec4 candidate = ${k};
          bvec4 nan = isnan(candidate);
          bvec4 replace = bvec4(
            vec4(${v}(candidate, bestValue)) * (vec4(1.0) - vec4(nan)));

          bestValue = vec4(replace.x  ? candidate.x : bestValue.x,
                           replace.y  ? candidate.y : bestValue.y,
                           replace.z  ? candidate.z : bestValue.z,
                           replace.w  ? candidate.w : bestValue.w);
          bestIndex = mix(bestIndex, vec4(inIdx), vec4(replace));
          srcIdx++;
        }
        setOutput(bestIndex);
      }
    `;
  }
};
function n2(e, t, n, s = null) {
  let r = t.shape[0], a = t.shape[1];
  s != null && (r = s.shape[0], a = s.shape[1]);
  let i = C.computeOptimalWindowSize(a), o = { windowSize: i, inSize: a, batchSize: r, outSize: Math.ceil(a / i) }, u = new Z8(o, n, s == null), l = [t];
  s != null && l.push(s);
  let c = e.runWebGLProgram(u, l, "int32");
  if (c.shape[1] === 1)
    return c;
  let p = n2(e, t, n, c);
  return e.disposeIntermediateTensorInfo(c), p;
}
function s2(e, t, n, s = null) {
  let r = s != null ? s.shape : t.shape, a = r[r.length - 1], i = C.computeOptimalWindowSize(a), o = new J8(r, i, n, s == null), u = s == null ? [t] : [t, s], l = e.runWebGLProgram(o, u, "int32");
  if (l.shape.length === t.shape.length) {
    let c = s2(e, t, n, l);
    return e.disposeIntermediateTensorInfo(l), c;
  }
  return l;
}
function r2(e, t, n, s) {
  let r = [n];
  if (C.assertAxesAreInnerMostDims("arg" + s.charAt(0).toUpperCase() + s.slice(1), r, t.shape.length), !K().getBool("WEBGL_PACK_REDUCE") || t.shape.length <= 2) {
    let a = [], i = e.texData.get(t.dataId), o = i !== null && i.isPacked, u = t;
    o && (u = e.unpackTensor(t), a.push(u));
    let [l, c] = C.computeOutAndReduceShapes(u.shape, r), p = w.sizeFromShape(c), d = he({ inputs: { x: u }, backend: e, attrs: { shape: [-1, p] } });
    a.push(d);
    let h = n2(e, d, s);
    a.push(h);
    let f = he({ inputs: { x: h }, backend: e, attrs: { shape: l } });
    return a.forEach((m) => e.disposeIntermediateTensorInfo(m)), f;
  }
  return s2(e, t, s);
}
function eY(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a } = s, i = w.parseAxisParam(a, r.shape), o = C.getAxesPermutation(i, r.shape.length), u = r, l = [];
  o != null && (u = _t({ inputs: { x: r }, backend: n, attrs: { perm: o } }), l.push(u), i = C.getInnerMostAxes(i.length, u.shape.length)), C.assertAxesAreInnerMostDims("argMax", [i[0]], u.shape.length);
  let c = r2(n, u, i[0], "max");
  return l.forEach((p) => n.disposeIntermediateTensorInfo(p)), c;
}
var tY = { kernelName: Ia, backendName: "webgl", kernelFunc: eY };
function nY(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a } = s, i = w.parseAxisParam(a, r.shape), o = C.getAxesPermutation(i, r.shape.length), u = r, l = [];
  o != null && (u = _t({ inputs: { x: r }, backend: n, attrs: { perm: o } }), l.push(u), i = C.getInnerMostAxes(i.length, u.shape.length)), C.assertAxesAreInnerMostDims("argMin", [i[0]], u.shape.length);
  let c = r2(n, u, i[0], "min");
  return l.forEach((p) => n.disposeIntermediateTensorInfo(p)), c;
}
var sY = { kernelName: dl, backendName: "webgl", kernelFunc: nY };
var rY = ss + `
  if (abs(x) > 1.) {
    return NAN;
  }
  return asin(x);
`;
var aY = Ke({ opSnippet: rY });
var iY = { kernelName: pl, backendName: "webgl", kernelFunc: aY };
var oY = ss + "return log(x + sqrt(x * x + 1.0));";
var uY = Ke({ opSnippet: oY });
var lY = { kernelName: hl, backendName: "webgl", kernelFunc: uY };
var cY = ss + `
  return atan(x);
`;
var dY = Ke({ opSnippet: cY });
var pY = { kernelName: fl, backendName: "webgl", kernelFunc: dY };
var hY = x8 + `
  return atan(a, b);
`;
var fY = `
  vec4 result = atan(a, b);
  vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
  ` + w8 + `
  return result;
`;
var mY = jt({ opSnippet: hY, packedOpSnippet: fY });
var gY = { kernelName: gl, backendName: "webgl", kernelFunc: mY };
var bY = ss + `
  if ((x < -1.0) || (x > 1.0)) return NAN;
return (log(1.0 + x) - log(1.0 - x)) / 2.0;`;
var yY = Ke({ opSnippet: bY });
var vY = { kernelName: ml, backendName: "webgl", kernelFunc: yY };
var al = class {
  constructor(e, t, n, s = false, r = false) {
    if (this.variableNames = ["x"], t === "avg" && n)
      throw new Error("Cannot compute positions for average pool.");
    let a = e.filterWidth, i = e.strideHeight, o = e.strideWidth, u = e.dilationHeight, l = e.dilationWidth, c = e.effectiveFilterHeight, p = e.effectiveFilterWidth, d = e.padInfo.top, h = e.padInfo.left;
    this.outputShape = e.outShape;
    let f = t === "avg", m = `((batch  * ${e.inHeight} + xR) * ${e.inWidth} + xC) * ${e.inChannels} + d`, g = `(xR * ${e.inWidth} + xC) * ${e.inChannels} + d`, b = "0.0";
    if (f || (b = "-1.0 / 1e-20"), n) {
      let $ = ">=";
      this.userCode = `
        const ivec2 strides = ivec2(${i}, ${o});
        const ivec2 pads = ivec2(${d}, ${h});

        void main() {
          ivec4 coords = getOutputCoords();
          int batch = coords[0];
          int d = coords[3];

          ivec2 xRCCorner = coords.yz * strides - pads;
          int xRCorner = xRCCorner.x;
          int xCCorner = xRCCorner.y;

          // max/min x(?, ?, d) to get y(yR, yC, d).
          // ? = to be determined
          float minMaxValue = 0.0;
          float minMaxValueFound = 0.0;
          int minMaxPosition = 0;
          float avgValue = 0.0;

          for (int wR = 0; wR < ${c};
              wR += ${u}) {
            int xR = xRCorner + wR;

            if (xR < 0 || xR >= ${e.inHeight}) {
              continue;
            }

            for (int wC = 0; wC < ${p};
                wC += ${l}) {
              int xC = xCCorner + wC;

              if (xC < 0 || xC >= ${e.inWidth}) {
                continue;
              }

              float value = getX(batch, xR, xC, d);

              // If a min / max value has already been found, use it. If not,
              // use the current value.
              float currMinMaxValue = mix(
                  value, minMaxValue, minMaxValueFound);
              if (value ${$} currMinMaxValue) {
                minMaxValue = value;
                minMaxValueFound = 1.0;
                minMaxPosition = ${s ? r ? m : g : `wR * ${p} + wC`};
              }
            }
          }
          setOutput(float(minMaxPosition));
        }
      `;
      return;
    }
    let y = "max", v = `${t}(${t}(${t}(minMaxValue[0], minMaxValue[1]), minMaxValue[2]), minMaxValue[3])`;
    t === "avg" && (v = "avgValue / count");
    let x = Math.floor(a / 4) * 4, k = a % 4, I = `
      if (${f}) {
        avgValue += dot(values, ones);
      } else {
        minMaxValue = ${y}(values, minMaxValue);
      }
    `;
    this.userCode = `
      const ivec2 strides = ivec2(${i}, ${o});
      const ivec2 pads = ivec2(${d}, ${h});
      const float initializationValue = ${b};
      const vec4 ones = vec4(1.0, 1.0, 1.0, 1.0);

      float count = 0.0;

      float getValue(int batch, int xR, int xC, int d) {
        if (xC < 0 || xC >= ${e.inWidth}) {
          return initializationValue;
        }
        count += 1.0;
        return getX(batch, xR, xC, d);
      }

      void main() {
        ivec4 coords = getOutputCoords();
        int batch = coords[0];
        int d = coords[3];

        ivec2 xRCCorner = coords.yz * strides - pads;
        int xRCorner = xRCCorner.x;
        int xCCorner = xRCCorner.y;

        // max/min x(?, ?, d) to get y(yR, yC, d).
        // ? = to be determined
        vec4 minMaxValue = vec4(${b});
        float avgValue = 0.0;
        count = 0.0;

        for (int wR = 0; wR < ${c};
            wR += ${u}) {
          int xR = xRCorner + wR;

          if (xR < 0 || xR >= ${e.inHeight}) {
            continue;
          }

          for (int wC = 0; wC < ${x}; wC += 4) {
            int xC = xCCorner + wC * ${l};

            vec4 values = vec4(
              getValue(batch, xR, xC, d),
              getValue(batch, xR, xC + ${l}, d),
              getValue(batch, xR, xC + 2 * ${l}, d),
              getValue(batch, xR, xC + 3 * ${l}, d)
            );

            ${I}
          }

          int xC = xCCorner + ${x};
          if (${k === 1}) {
            vec4 values = vec4(
              getValue(batch, xR, xC, d),
              initializationValue,
              initializationValue,
              initializationValue
            );

            ${I}
          } else if (${k === 2}) {
            vec4 values = vec4(
              getValue(batch, xR, xC, d),
              getValue(batch, xR, xC + ${l}, d),
              initializationValue,
              initializationValue
            );

            ${I}
          } else if (${k === 3}) {
            vec4 values = vec4(
              getValue(batch, xR, xC, d),
              getValue(batch, xR, xC + ${l}, d),
              getValue(batch, xR, xC + 2 * ${l}, d),
              initializationValue
            );

            ${I}
          }
        }
        setOutput(${v});
      }
    `;
  }
};
var Av = class {
  constructor(e, t, n, s = false, r = false) {
    if (this.variableNames = ["x"], t === "avg" && n)
      throw new Error("Cannot compute positions for average pool.");
    let a = e.filterWidth, i = e.strideDepth, o = e.strideHeight, u = e.strideWidth, l = e.dilationDepth, c = e.dilationHeight, p = e.dilationWidth, d = e.effectiveFilterDepth, h = e.effectiveFilterHeight, f = e.effectiveFilterWidth, m = e.padInfo.front, g = e.padInfo.top, b = e.padInfo.left;
    this.outputShape = e.outShape;
    let y = t === "avg", v = "0.0";
    if (y || (v = "-1.0 / 1e-20"), n) {
      let E = ">=";
      this.userCode = `
        const ivec3 strides =
            ivec3(${i}, ${o}, ${u});
        const ivec3 pads = ivec3(${m}, ${g}, ${b});

        void main() {
          ivec5 coords = getOutputCoords();
          int batch = coords.x;
          int ch = coords.u;

          ivec3 xCorner = ivec3(coords.y, coords.z, coords.w) * strides - pads;
          int xDCorner = xCorner.x;
          int xRCorner = xCorner.y;
          int xCCorner = xCorner.z;

          // max/min x(?, ?, ?, ch) to get y(yD, yR, yC, ch).
          // ? = to be determined
          float minMaxValue = 0.0;
          float minMaxValueFound = 0.0;
          int minMaxPosition = 0;

          for (int wD = 0; wD < ${d};
              wD += ${l}) {
            int xD = xDCorner + wD;

            if (xD < 0 || xD >= ${e.inDepth}) {
              continue;
            }

            for (int wR = 0; wR < ${h};
                wR += ${c}) {
              int xR = xRCorner + wR;

              if (xR < 0 || xR >= ${e.inHeight}) {
                continue;
              }

              for (int wC = 0; wC < ${f};
                  wC += ${p}) {
                int xC = xCCorner + wC;

                if (xC < 0 || xC >= ${e.inWidth}) {
                  continue;
                }

                float value = getX(batch, xD, xR, xC, ch);

                // If a min / max value has already been found, use it. If not,
                // use the current value.
                float currMinMaxValue = mix(
                    value, minMaxValue, minMaxValueFound);
                if (value ${E} currMinMaxValue) {
                  minMaxValue = value;
                  minMaxValueFound = 1.0;
                  minMaxPosition = ${s ? r ? `(((batch * ${e.inDepth} + xD) * ${e.inHeight} + xR) * ${e.inWidth} + xC) * ${e.inChannels} + ch` : `((xD * ${e.inHeight} + xR) * ${e.inWidth} + xC) * ${e.inChannels} + ch` : `wD * ${h} * ${f} +
                      wR * ${f} + wC`};
                }
              }
            }
          }
          setOutput(float(minMaxPosition));
        }
      `;
      return;
    }
    let x = "max", k = `${t}(${t}(${t}(minMaxValue[0], minMaxValue[1]), minMaxValue[2]), minMaxValue[3])`;
    t === "avg" && (k = "avgValue / count");
    let I = Math.floor(a / 4) * 4, $ = a % 4, R = `
      if (${y}) {
        avgValue += dot(values, ones);
      } else {
        minMaxValue = ${x}(values, minMaxValue);
      }
    `;
    this.userCode = `
      const ivec3 strides =
        ivec3(${i}, ${o}, ${u});
      const ivec3 pads = ivec3(${m}, ${g}, ${b});
      const float initializationValue = ${v};
      const vec4 ones = vec4(1.0, 1.0, 1.0, 1.0);

      float count = 0.0;

      float getValue(int batch, int xD, int xR, int xC, int ch) {
        if (xC < 0 || xC >= ${e.inWidth}) {
          return initializationValue;
        }
        count += 1.0;
        return getX(batch, xD, xR, xC, ch);
      }

      void main() {
        ivec5 coords = getOutputCoords();
        int batch = coords.x;
        int ch = coords.u;

        ivec3 xCorner = ivec3(coords.y, coords.z, coords.w) * strides - pads;
        int xDCorner = xCorner.x;
        int xRCorner = xCorner.y;
        int xCCorner = xCorner.z;

        // max/min x(?, ?, ?, d) to get y(yD, yR, yC, ch).
        // ? = to be determined
        vec4 minMaxValue = vec4(${v});
        float avgValue = 0.0;
        count = 0.0;

        for (int wD = 0; wD < ${d};
            wD += ${l}) {
          int xD = xDCorner + wD;

          if (xD < 0 || xD >= ${e.inDepth}) {
            continue;
          }

          for (int wR = 0; wR < ${h};
            wR += ${c}) {
            int xR = xRCorner + wR;

            if (xR < 0 || xR >= ${e.inHeight}) {
              continue;
            }

            for (int wC = 0; wC < ${I}; wC += 4) {
              int xC = xCCorner + wC * ${p};

              vec4 values = vec4(
                getValue(batch, xD, xR, xC, ch),
                getValue(batch, xD, xR, xC + ${p}, ch),
                getValue(batch, xD, xR, xC + 2 * ${p}, ch),
                getValue(batch, xD, xR, xC + 3 * ${p}, ch)
              );

              ${R}
            }

            int xC = xCCorner + ${I};
            if (${$ === 1}) {
              vec4 values = vec4(
                getValue(batch, xD, xR, xC, ch),
                initializationValue,
                initializationValue,
                initializationValue
              );

              ${R}
            } else if (${$ === 2}) {
              vec4 values = vec4(
                getValue(batch, xD, xR, xC, ch),
                getValue(batch, xD, xR, xC + ${p}, ch),
                initializationValue,
                initializationValue
              );

              ${R}
            } else if (${$ === 3}) {
              vec4 values = vec4(
                getValue(batch, xD, xR, xC, ch),
                getValue(batch, xD, xR, xC + ${p}, ch),
                getValue(batch, xD, xR, xC + 2 * ${p}, ch),
                initializationValue
              );

              ${R}
            }
          }
          setOutput(${k});
        }
      }
    `;
  }
};
function xY(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t;
  iu(r, "avgPool");
  let { filterSize: a, strides: i, pad: o, dimRoundingMode: u } = s, l = 1;
  w.assert(C.eitherStridesOrDilationsAreOne(i, l), () => `Error in avgPool: Either strides or dilations must be 1. Got strides ${i} and dilations '${l}'`);
  let c = C.computePool2DInfo(r.shape, a, i, l, o, u);
  if (c.filterWidth === 1 && c.filterHeight === 1 && w.arraysEqual(c.inShape, c.outShape))
    return Rn({ inputs: { x: r }, backend: n });
  let p = new al(c, "avg", false);
  return n.runWebGLProgram(p, [r], "float32");
}
var wY = { kernelName: Ca, backendName: "webgl", kernelFunc: xY };
function kY(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { filterSize: a, strides: i, pad: o, dimRoundingMode: u, dataFormat: l } = s, c = [1, 1, 1], p = C.computePool3DInfo(r.shape, a, i, c, o, u, l), d = new Av(p, "avg", false);
  return n.runWebGLProgram(d, [r], "float32");
}
var SY = { kernelName: Jd, backendName: "webgl", kernelFunc: kY };
var IY = class {
  constructor(e) {
    this.variableNames = ["dy"], this.outputShape = e.inShape;
    let t = e.filterHeight, n = e.filterWidth, s = e.strideHeight, r = e.strideWidth, a = e.dilationHeight, i = e.dilationWidth, o = e.effectiveFilterHeight, u = e.effectiveFilterWidth, l = o - 1 - e.padInfo.top, c = u - 1 - e.padInfo.left, p = 1 / (t * n);
    this.userCode = `
      const ivec2 pads = ivec2(${l}, ${c});
      const float avgMultiplier = float(${p});

      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int d = coords[3];

        ivec2 dyRCCorner = coords.yz - pads;
        int dyRCorner = dyRCCorner.x;
        int dyCCorner = dyRCCorner.y;

        // Convolve dy(?, ?, d) with pos mask(:, :, d) to get dx(xR, xC, d).
        // ? = to be determined. : = across all values in that axis.
        float dotProd = 0.0;
        for (int wR = 0; wR < ${o};
            wR += ${a}) {
          float dyR = float(dyRCorner + wR) / ${s}.0;

          if (dyR < 0.0 || dyR >= ${e.outHeight}.0 || fract(dyR) > 0.0) {
            continue;
          }
          int idyR = int(dyR);

          for (int wC = 0; wC < ${u};
            wC+= ${i}) {
            float dyC = float(dyCCorner + wC) / ${r}.0;

            if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
                fract(dyC) > 0.0) {
              continue;
            }
            int idyC = int(dyC);

            float dyValue = getDy(b, idyR, idyC, d);

            dotProd += dyValue * avgMultiplier;
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
var CY = class {
  constructor(e) {
    this.variableNames = ["dy"], this.outputShape = e.inShape;
    let t = e.filterDepth, n = e.filterHeight, s = e.filterWidth, r = e.strideDepth, a = e.strideHeight, i = e.strideWidth, o = e.dilationDepth, u = e.dilationHeight, l = e.dilationWidth, c = e.effectiveFilterDepth, p = e.effectiveFilterHeight, d = e.effectiveFilterWidth, h = c - 1 - e.padInfo.front, f = p - 1 - e.padInfo.top, m = d - 1 - e.padInfo.left, g = 1 / (t * n * s);
    this.userCode = `
      const ivec3 pads = ivec3(${h}, ${f}, ${m});
      const float avgMultiplier = float(${g});

      void main() {
        ivec5 coords = getOutputCoords();
        int batch = coords.x;
        int ch = coords.u;

        ivec3 dyCorner = ivec3(coords.y, coords.z, coords.w) - pads;
        int dyDCorner = dyCorner.x;
        int dyRCorner = dyCorner.y;
        int dyCCorner = dyCorner.z;

        // Convolve dy(?, ?, ?, d) with pos mask(:, :, :, ch) to get
        // dx(xD, xR, xC, ch).
        // ? = to be determined. : = across all values in that axis.
        float dotProd = 0.0;

        for (int wD = 0; wD < ${c};
            wD += ${o}) {
          float dyD = float(dyDCorner + wD) / ${r}.0;

          if (dyD < 0.0 || dyD >= ${e.outDepth}.0 || fract(dyD) > 0.0) {
            continue;
          }
          int idyD = int(dyD);

          for (int wR = 0; wR < ${p};
              wR += ${u}) {
            float dyR = float(dyRCorner + wR) / ${a}.0;

            if (dyR < 0.0 || dyR >= ${e.outHeight}.0 ||
                fract(dyR) > 0.0) {
              continue;
            }
            int idyR = int(dyR);

            for (int wC = 0; wC < ${d};
                wC += ${l}) {
              float dyC = float(dyCCorner + wC) / ${i}.0;

              if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
                  fract(dyC) > 0.0) {
                continue;
              }
              int idyC = int(dyC);

              float dyValue = getDy(batch, idyD, idyR, idyC, ch);

              dotProd += dyValue * avgMultiplier;
            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
function NY(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, input: a } = t, i = a, { filterSize: o, strides: u, pad: l, dimRoundingMode: c } = s, p = [1, 1, 1], d = C.computePool3DInfo(i.shape, o, u, p, l, c), h = new CY(d);
  return n.runWebGLProgram(h, [r], i.dtype);
}
var TY = { kernelName: hg, backendName: "webgl", kernelFunc: NY };
function $Y(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, input: a } = t, i = a;
  iu([r, a], "avgPoolGrad");
  let { filterSize: o, strides: u, pad: l } = s, c = C.computePool2DInfo(i.shape, o, u, 1, l), p = new IY(c);
  return n.runWebGLProgram(p, [r], i.dtype);
}
var _Y = { kernelName: pg, backendName: "webgl", kernelFunc: $Y };
function AY(e) {
  let { inputs: t, backend: n, attrs: s } = e, { a: r, b: a } = t, { transposeA: i, transposeB: o } = s;
  return Gd({ a: r, b: a, transposeA: i, transposeB: o, backend: n });
}
var EY = { kernelName: Na, backendName: "webgl", kernelFunc: AY };
var RY = class {
  constructor(e, t, n, s, r, a) {
    this.outputShape = [], this.variableNames = ["x", "mean", "variance"], C.assertAndGetBroadcastShape(e, t), C.assertAndGetBroadcastShape(e, n);
    let i = "0.0";
    s != null && (C.assertAndGetBroadcastShape(e, s), this.variableNames.push("offset"), i = "getOffsetAtOutCoords()");
    let o = "1.0";
    r != null && (C.assertAndGetBroadcastShape(e, r), this.variableNames.push("scale"), o = "getScaleAtOutCoords()"), this.outputShape = e, this.userCode = `
      void main() {
        float x = getXAtOutCoords();
        float mean = getMeanAtOutCoords();
        float variance = getVarianceAtOutCoords();
        float offset = ${i};
        float scale = ${o};
        float inv = scale * inversesqrt(variance + float(${a}));
        setOutput(dot(vec3(x, -mean, offset), vec3(inv, inv, 1)));
      }
    `;
  }
};
var DY = class {
  constructor(e, t, n, s, r, a) {
    this.packedInputs = true, this.packedOutput = true, this.variableNames = ["x", "mean", "variance"], C.assertAndGetBroadcastShape(e, t), C.assertAndGetBroadcastShape(e, n);
    let i = "vec4(0.0)";
    s != null && (C.assertAndGetBroadcastShape(e, s), this.variableNames.push("offset"), i = "getOffsetAtOutCoords()");
    let o = "vec4(1.0)";
    r != null && (C.assertAndGetBroadcastShape(e, r), this.variableNames.push("scale"), o = "getScaleAtOutCoords()"), this.outputShape = e, this.userCode = `
      void main() {
        vec4 offset = ${i};
        vec4 scale = ${o};

        vec4 x = getXAtOutCoords();
        vec4 mean = getMeanAtOutCoords();
        vec4 variance = getVarianceAtOutCoords();

        vec4 inv = scale * inversesqrt(variance + vec4(${a}));

        setOutput((x - mean) * inv + offset);
      }
    `;
  }
};
var FY = ({ inputs: e, backend: t, attrs: n }) => {
  let { x: s, mean: r, variance: a, offset: i, scale: o } = e;
  w.assert(r.shape.length === a.shape.length, () => "Batch normalization gradient requires mean and variance to have equal ranks."), w.assert(i == null || r.shape.length === i.shape.length, () => "Batch normalization gradient requires mean and offset to have equal ranks."), w.assert(o == null || r.shape.length === o.shape.length, () => "Batch normalization gradient requires mean and scale to have equal ranks.");
  let { varianceEpsilon: u } = n;
  u == null && (u = 1e-3);
  let l = [s, r, a], c = null;
  i != null && (c = i.shape, l.push(i));
  let p = null;
  o != null && (p = o.shape, l.push(o));
  let d = K().getBool("WEBGL_PACK_NORMALIZATION") ? new DY(s.shape, r.shape, a.shape, c, p, u) : new RY(s.shape, r.shape, a.shape, c, p, u);
  return t.runWebGLProgram(d, l, l[0].dtype);
};
var OY = { kernelName: Ba, backendName: "webgl", kernelFunc: FY };
var PY = class {
  constructor(e) {
    this.variableNames = ["source"], this.outputShape = e, this.rank = e.length;
    let t = ot(this.rank);
    this.customUniforms = [{ name: "start", arrayIndex: this.rank, type: "int" }];
    let n = zY(this.rank), s, r = e.map((a, i) => `sourceLoc.${Qm[i]} = start[${i}] + coords.${Qm[i]};`);
    s = `
        ${t} sourceLoc;
        ${t} coords = getOutputCoords();
        ${r.join(`
`)}
      `, this.userCode = `
      void main() {
        ${s}
        setOutput(getSource(${n}));
      }
    `;
  }
};
var Qm = ["x", "y", "z", "w", "u", "v"];
function zY(e) {
  if (e === 1)
    return "sourceLoc";
  if (e <= 6)
    return Qm.slice(0, e).map((t) => "sourceLoc." + t).join(",");
  throw Error(`Slicing for rank ${e} is not yet supported`);
}
var MY = class {
  constructor(e) {
    this.variableNames = ["source"], this.packedInputs = true, this.packedOutput = true, this.outputShape = e, this.rank = e.length, this.customUniforms = [{ name: "start", arrayIndex: this.rank, type: "int" }];
    let t = ot(this.rank), n = ln("coords", this.rank), s = ln("sourceLoc", this.rank), r = this.rank === 1 ? "sourceLoc" : `vec2(${s.slice(-2).join()})`, a = `getChannel(getSource(${s.join()}), ${r})`, i = `
      result.x = ${a};
      if (++${n[this.rank - 1]} < ${e[this.rank - 1]}) {
        ++${s[this.rank - 1]};
        result.y = ${a};
        --${s[this.rank - 1]};
      }
    `, o = this.rank === 1 ? "" : `
      --${n[this.rank - 1]};
      if (++${n[this.rank - 2]} < ${e[this.rank - 2]}) {
        ++${s[this.rank - 2]};
        result.z = ${a};
        if (++${n[this.rank - 1]} < ${e[this.rank - 1]}) {
          ++${s[this.rank - 1]};
          result.w = ${a};
        }
      }
    `, u = this.rank <= 4 ? `sourceLoc = coords +
            ${t}(${e.map((l, c) => `start[${c}]`).join()});` : e.map((l, c) => `${s[c]} = ${n[c]} + start[${c}];`).join(`
`);
    this.userCode = `
      void main() {
        ${t} coords = getOutputCoords();
        ${t} sourceLoc;
        ${u}
        vec4 result = vec4(0.);
        ${i}
        ${o}
        setOutput(result);
      }
    `;
  }
};
function LY(e, t, n, s) {
  let r = s.texData.get(e.dataId), a = s.makeTensorInfo(n, e.dtype), i = s.texData.get(a.dataId);
  Object.assign(i, r), i.refCount = 1, i.shape = n, i.dtype = e.dtype;
  let o = wt.computeFlatOffset(t, w.computeStrides(e.shape));
  r.slice && (o += r.slice.flatOffset), i.slice = { flatOffset: o, origDataId: r.slice && r.slice.origDataId || e.dataId };
  let u = s.dataRefCount.get(i.slice.origDataId) || 1;
  return s.dataRefCount.set(i.slice.origDataId, u + 1), a;
}
function pu(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { begin: a, size: i } = s, [o, u] = wt.parseSliceParams(r, a, i);
  if (wt.assertParamsValid(r, o, u), w.sizeFromShape(u) === 0)
    return n.makeTensorInfo(u, r.dtype, []);
  if (n.shouldExecuteOnCPU([r]) || r.dtype === "string") {
    let p = n.texData.get(r.dataId), d = _X(p.values, o, u, r.shape, r.dtype);
    return n.makeTensorInfo(u, r.dtype, d);
  }
  let { isPacked: l } = n.texData.get(r.dataId), c = wt.isSliceContinous(r.shape, o, u);
  if (l || !c) {
    let p = K().getBool("WEBGL_PACK_ARRAY_OPERATIONS") ? new MY(u) : new PY(u), d = [o];
    return n.runWebGLProgram(p, [r], r.dtype, d);
  }
  return n.uploadToGPU(r.dataId), LY(r, o, u, n);
}
var BY = { kernelName: Bo, backendName: "webgl", kernelFunc: pu };
var VY = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockShape: a, crops: i } = s;
  w.assert(r.shape.length <= 4, () => "batchToSpaceND for rank > 4 with a WebGL backend not implemented yet");
  let o = a.reduce((y, v) => y * v), u = C.getReshaped(r.shape, a, o), l = C.getPermuted(u.length, a.length), c = C.getReshapedPermuted(r.shape, a, o), p = C.getSliceBeginCoords(i, a.length), d = C.getSliceSize(c, i, a.length), h = [], f = he({ inputs: { x: r }, backend: n, attrs: { shape: u } }), m = _t({ inputs: { x: f }, backend: n, attrs: { perm: l } }), g = he({ inputs: { x: m }, backend: n, attrs: { shape: c } }), b = pu({ inputs: { x: g }, backend: n, attrs: { begin: p, size: d } });
  return h.push(f), h.push(m), h.push(g), h.forEach((y) => n.disposeIntermediateTensorInfo(y)), b;
};
var WY = { kernelName: ho, backendName: "webgl", kernelFunc: VY };
function UY(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, weights: a } = t, { size: i } = s, o = n.readSync(r.dataId), u = n.readSync(a.dataId), l = W1(o, u, a.dtype, a.shape, i);
  return n.makeTensorInfo([i], a.dtype, l);
}
var GY = { kernelName: fg, backendName: "webgl", kernelFunc: UY };
function HY(e) {
  let { inputs: t, backend: n } = e, { s0: s, s1: r } = t, a = n.readSync(s.dataId), i = n.readSync(r.dataId), o = C.assertAndGetBroadcastShape(Array.from(a), Array.from(i));
  return n.makeTensorInfo([o.length], "int32", Int32Array.from(o));
}
var qY = { kernelName: mg, backendName: "webgl", kernelFunc: HY };
var jY = "return float(a != b);";
var a2 = jt({ opSnippet: jY, cpuKernelImpl: SX, dtype: "bool" });
var KY = { kernelName: _o, backendName: "webgl", kernelFunc: a2 };
function tc(e) {
  let { inputs: t, backend: n } = e, { input: s } = t, r = n.texData.get(s.dataId);
  return Rn({ inputs: { x: r.complexTensorInfos.real }, backend: n });
}
var XY = { kernelName: lp, backendName: "webgl", kernelFunc: tc };
var YY = "return float(int(x));";
function QY(e, t) {
  let n = new Gs(e.shape, YY), s = t.runWebGLProgram(n, [e], "int32");
  return { dataId: s.dataId, shape: s.shape, dtype: s.dtype };
}
function Zm(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { dtype: a } = s;
  if (a === "complex64") {
    if (r.dtype === "complex64")
      return Rn({ inputs: { x: r }, backend: n });
    let i = $t(r.shape), o = Zm({ inputs: { x: r }, backend: n, attrs: { dtype: "float32" } }), u = Rr({ inputs: { real: o, imag: i }, backend: n });
    return i.dispose(), n.disposeIntermediateTensorInfo(o), u;
  }
  if (r.dtype === "complex64") {
    let i = tc({ inputs: { input: r }, backend: n }), o = Zm({ inputs: { x: i }, backend: n, attrs: { dtype: a } });
    return n.disposeIntermediateTensorInfo(i), o;
  }
  if (!w.hasEncodingLoss(r.dtype, a)) {
    let i = Rn({ inputs: { x: r }, backend: n });
    return { dataId: i.dataId, shape: i.shape, dtype: a };
  }
  if (a === "int32")
    return QY(r, n);
  if (a === "bool") {
    let i = n.makeTensorInfo([], "bool", w.getTypedArrayFromDType("bool", 1)), u = a2({ inputs: { a: r, b: i }, backend: n });
    return n.disposeIntermediateTensorInfo(i), u;
  }
  throw new Error(`Error in Cast: failed to cast ${r.dtype} to ${a}`);
}
var ZY = { kernelName: Ta, backendName: "webgl", kernelFunc: Zm };
var ww = "return ceil(x);";
var JY = Ke({ opSnippet: ww, packedOpSnippet: ww, cpuKernelImpl: rX });
var e9 = { kernelName: $a, backendName: "webgl", kernelFunc: JY };
var t9 = class {
  constructor(e) {
    this.variableNames = ["A"], this.customUniforms = [{ name: "minVal", type: "float" }, { name: "maxVal", type: "float" }], this.outputShape = e, this.userCode = `

      void main() {
        float value = getAAtOutCoords();
        if (isnan(value)) {
          setOutput(value);
          return;
        }

        setOutput(clamp(value, minVal, maxVal));
      }
    `;
  }
};
var n9 = class {
  constructor(e) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true, this.customUniforms = [{ name: "minVal", type: "float" }, { name: "maxVal", type: "float" }], this.outputShape = e, this.userCode = `
      void main() {
        vec4 value = getAAtOutCoords();

        if (any(isnan(value))) {
          setOutput(value);
          return;
        }

        setOutput(clamp(value, vec4(minVal), vec4(maxVal)));
      }
    `;
  }
};
function s9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { clipValueMin: a, clipValueMax: i } = s, o;
  K().getBool("WEBGL_PACK_CLIP") ? o = new n9(r.shape) : o = new t9(r.shape);
  let u = [[a], [i]];
  return n.runWebGLProgram(o, [r], r.dtype, u);
}
var r9 = { kernelName: Ir, backendName: "webgl", kernelFunc: s9 };
var a9 = class {
  constructor(e) {
    this.variableNames = ["real", "imag"], this.outputShape = e, this.userCode = `
      void main() {
        float re = abs(getRealAtOutCoords());
        float im = abs(getImagAtOutCoords());
        float mx = max(re, im);

        // sadly the length function in glsl is not underflow-safe
        // (at least not on Intel GPUs). So the safe solution is
        // to ensure underflow-safety in all cases.
        setOutput(
          mx == 0.0 ? 0.0 : mx * length(vec2(1, min(re, im)/mx))
        );
      }
    `;
  }
};
function kw(e, t) {
  return { dataId: t.dataId, dtype: t.dtype, shape: e.shape };
}
function i9(e) {
  let { inputs: t, backend: n } = e, { x: s } = t, r = n.texData.get(s.dataId), a = new a9(s.shape), i = [kw(s, r.complexTensorInfos.real), kw(s, r.complexTensorInfos.imag)];
  return n.runWebGLProgram(a, i, i[0].dtype);
}
var o9 = { kernelName: tp, backendName: "webgl", kernelFunc: i9 };
var u9 = class {
  constructor(e) {
    this.outputShape = [], this.outputShape = C.computeOutShape(e, 1), this.variableNames = e.map((a, i) => `T${i}`);
    let t = new Array(e.length - 1);
    t[0] = e[0][1];
    for (let a = 1; a < t.length; a++)
      t[a] = t[a - 1] + e[a][1];
    let n = [`if (yC < ${t[0]}) setOutput(getT0(yR, yC));`];
    for (let a = 1; a < t.length; a++) {
      let i = t[a - 1];
      n.push(`else if (yC < ${t[a]}) setOutput(getT${a}(yR, yC-${i}));`);
    }
    let s = t.length, r = t[t.length - 1];
    n.push(`else setOutput(getT${s}(yR, yC-${r}));`), this.userCode = `
      void main() {
        ivec2 coords = getOutputCoords();
        int yR = coords.x;
        int yC = coords.y;

        ${n.join(`
        `)}
      }
    `;
  }
};
var l9 = class {
  constructor(e, t) {
    this.packedInputs = true, this.packedOutput = true, this.outputShape = [], this.outputShape = C.computeOutShape(e, t);
    let n = this.outputShape, s = n.length, r = ot(s), a = ln("coords", s), i = ["x", "y", "z", "w", "u", "v"].slice(0, s);
    this.variableNames = e.map((f, m) => `T${m}`);
    let o = new Array(e.length - 1);
    o[0] = e[0][t];
    for (let f = 1; f < o.length; f++)
      o[f] = o[f - 1] + e[f][t];
    let u = i[t], l = i.slice(-2), c = i.join(), p = `if (${u} < ${o[0]}) {
        return getChannel(
            getT0(${c}), vec2(${l.join()}));
        }`;
    for (let f = 1; f < o.length; f++) {
      let m = o[f - 1];
      p += `
        if (${u} < ${o[f]}  && ${u} >= ${o[f - 1]}) {
          return getChannel(
            getT${f}(${nd(i, u, m)}),
            vec2(${nd(l, u, m)}));
        }`;
    }
    let d = o.length, h = o[o.length - 1];
    p += `
        return getChannel(
          getT${d}(${nd(i, u, h)}),
          vec2(${nd(l, u, h)}));`, this.userCode = `
      float getValue(${i.map((f) => "int " + f)}) {
        ${p}
      }

      void main() {
        ${r} coords = getOutputCoords();
        vec4 result = vec4(getValue(${a}), 0., 0., 0.);

        ${a[s - 1]} = ${a[s - 1]} + 1;
        if (${a[s - 1]} < ${n[s - 1]}) {
          result.g = getValue(${a});
        }

        ${a[s - 2]} = ${a[s - 2]} + 1;
        if (${a[s - 2]} < ${n[s - 2]}) {
          result.a = getValue(${a});
        }

        ${a[s - 1]} = ${a[s - 1]} - 1;
        if (${a[s - 2]} < ${n[s - 2]} &&
            ${a[s - 1]} < ${n[s - 1]}) {
          result.b = getValue(${a});
        }
        setOutput(result);
      }
    `;
  }
};
function nd(e, t, n) {
  let s = e.indexOf(t);
  return e.map((a, i) => i === s ? `${a} - ${n}` : a).join();
}
function rh(e) {
  let { inputs: t, backend: n } = e, { input: s } = t, r = n.texData.get(s.dataId);
  return Rn({ inputs: { x: r.complexTensorInfos.imag }, backend: n });
}
var c9 = { kernelName: ap, backendName: "webgl", kernelFunc: rh };
function ji(e, t, n) {
  let s = e[0].dtype;
  if (s === "complex64") {
    let c = e.map((m) => tc({ inputs: { input: m }, backend: n })), p = e.map((m) => rh({ inputs: { input: m }, backend: n })), d = ji(c, t, n), h = ji(p, t, n), f = Rr({ inputs: { real: d, imag: h }, backend: n });
    return c.forEach((m) => n.disposeIntermediateTensorInfo(m)), p.forEach((m) => n.disposeIntermediateTensorInfo(m)), n.disposeIntermediateTensorInfo(d), n.disposeIntermediateTensorInfo(h), f;
  }
  let r = n.shouldExecuteOnCPU(e);
  if (s === "string" && (r = true), r) {
    let c = e.map((b) => {
      let y = w.sizeFromShape(b.shape.slice(t));
      return he({ inputs: { x: b }, backend: n, attrs: { shape: [-1, y] } });
    }), p = c.map((b) => ({ vals: n.readSync(b.dataId), shape: b.shape })), d = C.computeOutShape(c.map((b) => b.shape), 1), h = c[0].shape[0] === 1, f = aX(p, d, s, h), m = C.computeOutShape(e.map((b) => b.shape), t), g = n.makeTensorInfo(m, s, f);
    return c.forEach((b) => n.disposeIntermediateTensorInfo(b)), g;
  }
  if (e.length > K().getNumber("WEBGL_MAX_TEXTURES_IN_SHADER")) {
    let c = Math.floor(e.length / 2), p = ji(e.slice(0, c), t, n), d = ji(e.slice(c), t, n), h = ji([p, d], t, n);
    return n.disposeIntermediateTensorInfo(p), n.disposeIntermediateTensorInfo(d), h;
  }
  if (K().getBool("WEBGL_PACK_ARRAY_OPERATIONS") && e[0].shape.length > 1) {
    let c = new l9(e.map((p) => p.shape), t);
    return n.runWebGLProgram(c, e, s);
  }
  let { tensors2D: a, outShape: i } = d9(e, t, n), o = new u9(a.map((c) => c.shape)), u = n.runWebGLProgram(o, a, s);
  a.forEach((c) => n.disposeIntermediateTensorInfo(c));
  let l = he({ inputs: { x: u }, attrs: { shape: i }, backend: n });
  return n.disposeIntermediateTensorInfo(u), l;
}
function d9(e, t, n) {
  let s = C.computeOutShape(e.map((a) => a.shape), t);
  return { tensors2D: e.map((a) => he({ inputs: { x: a }, attrs: { shape: [-1, w.sizeFromShape(a.shape.slice(t))] }, backend: n })), outShape: s };
}
function i2(e) {
  let { inputs: t, backend: n, attrs: s } = e, { axis: r } = s, a = w.parseAxisParam(r, t[0].shape)[0], i = C.computeOutShape(t.map((l) => l.shape), a);
  if (w.sizeFromShape(i) === 0)
    return n.makeTensorInfo(i, t[0].dtype, []);
  let o = t.filter((l) => w.sizeFromShape(l.shape) > 0);
  if (o.length === 1)
    return Rn({ inputs: { x: o[0] }, backend: n });
  let u = o.map((l) => l.shape);
  return C.assertParamsConsistent(u, a), ji(o, a, n);
}
var p9 = { kernelName: fo, backendName: "webgl", kernelFunc: i2 };
var o2 = class {
  constructor(e, t = false, n = null, s = false, r = false) {
    this.variableNames = ["x", "W"], this.outputShape = e.outShape;
    let a = e.padInfo.top, i = e.padInfo.left, o = e.strideHeight, u = e.strideWidth, l = e.dilationHeight, c = e.dilationWidth, p = e.filterHeight, d = e.filterWidth, h = Math.floor(e.inChannels / 4) * 4, f = e.inChannels % 4, m = e.dataFormat === "channelsLast", g = m ? 1 : 2, b = m ? 2 : 3, y = m ? 3 : 1, v = "", x = "";
    n && (s ? v = `float activation(float a) {
          float b = getPreluActivationWeightsAtOutCoords();
          ${n}
        }` : r ? v = `float activation(float a) {
          float b = getLeakyreluAlphaAtOutCoords();
          ${n}
        }` : v = `
          float activation(float x) {
            ${n}
          }
        `, x = "result = activation(result);");
    let k = t ? "result += getBiasAtOutCoords();" : "";
    t && this.variableNames.push("bias"), s && this.variableNames.push("preluActivationWeights"), r && this.variableNames.push("leakyreluAlpha"), this.userCode = `
      ${v}

      const ivec2 strides = ivec2(${o}, ${u});
      const ivec2 pads = ivec2(${a}, ${i});

      void main() {
        ivec4 coords = getOutputCoords();
        int batch = coords[0];
        int d2 = coords[${y}];

        ivec2 xRCCorner =
            ivec2(coords[${g}], coords[${b}]) * strides - pads;
        int xRCorner = xRCCorner.x;
        int xCCorner = xRCCorner.y;

        // Convolve x(?, ?, d1) with w(:, :, d1, d2) to get y(yR, yC, d2).
        // ? = to be determined. : = across all values in that axis.
        float dotProd = 0.0;
        for (int wR = 0; wR < ${p}; wR++) {
          int xR = xRCorner + wR * ${l};

          if (xR < 0 || xR >= ${e.inHeight}) {
            continue;
          }

          for (int wC = 0; wC < ${d}; wC++) {
            int xC = xCCorner + wC * ${c};

            if (xC < 0 || xC >= ${e.inWidth}) {
              continue;
            }

            for (int d1 = 0; d1 < ${h}; d1 += 4) {
              vec4 wValues = vec4(
                getW(wR, wC, d1, d2),
                getW(wR, wC, d1 + 1, d2),
                getW(wR, wC, d1 + 2, d2),
                getW(wR, wC, d1 + 3, d2)
              );

              if (${m}) {
                vec4 xValues = vec4(
                  getX(batch, xR, xC, d1),
                  getX(batch, xR, xC, d1 + 1),
                  getX(batch, xR, xC, d1 + 2),
                  getX(batch, xR, xC, d1 + 3)
                );
                dotProd += dot(xValues, wValues);
              } else {
                vec4 xValues = vec4(
                  getX(batch, d1, xR, xC),
                  getX(batch, d1 + 1, xR, xC),
                  getX(batch, d1 + 2, xR, xC),
                  getX(batch, d1 + 3, xR, xC)
                );
                dotProd += dot(xValues, wValues);
              }
            }

            if (${f === 1}) {

              if (${m}) {
                dotProd +=
                    getX(batch, xR, xC, ${h}) *
                    getW(wR, wC, ${h}, d2);
              } else {
                dotProd +=
                    getX(batch, ${h}, xR, xC) *
                    getW(wR, wC, ${h}, d2);
              }

            } else if (${f === 2}) {
              vec2 wValues = vec2(
                getW(wR, wC, ${h}, d2),
                getW(wR, wC, ${h} + 1, d2)
              );

              if (${m}) {
                vec2 xValues = vec2(
                  getX(batch, xR, xC, ${h}),
                  getX(batch, xR, xC, ${h} + 1)
                );
                dotProd += dot(xValues, wValues);
              } else {
                vec2 xValues = vec2(
                  getX(batch, ${h}, xR, xC),
                  getX(batch, ${h} + 1, xR, xC)
                );
                dotProd += dot(xValues, wValues);
              }

            } else if (${f === 3}) {
              vec3 wValues = vec3(
                getW(wR, wC, ${h}, d2),
                getW(wR, wC, ${h} + 1, d2),
                getW(wR, wC, ${h} + 2, d2)
              );

              if (${m}) {
                vec3 xValues = vec3(
                  getX(batch, xR, xC, ${h}),
                  getX(batch, xR, xC, ${h} + 1),
                  getX(batch, xR, xC, ${h} + 2)
                );
                dotProd += dot(xValues, wValues);
              } else {
                vec3 xValues = vec3(
                  getX(batch, ${h}, xR, xC),
                  getX(batch, ${h} + 1, xR, xC),
                  getX(batch, ${h} + 2, xR, xC)
                );
                dotProd += dot(xValues, wValues);
              }

            }
          }
        }

        float result = dotProd;
        ${k}
        ${x}
        setOutput(result);
      }
    `;
  }
};
var h9 = class {
  constructor(e) {
    this.variableNames = ["x", "W"], this.outputShape = e.outShape;
    let t = e.padInfo.front, n = e.padInfo.top, s = e.padInfo.left, r = e.strideDepth, a = e.strideHeight, i = e.strideWidth, o = e.dilationDepth, u = e.dilationHeight, l = e.dilationWidth, c = e.filterDepth, p = e.filterHeight, d = e.filterWidth, h = Math.floor(e.inChannels / 4) * 4, f = e.inChannels % 4;
    this.userCode = `
      const ivec3 strides = ivec3(${r}, ${a}, ${i});
      const ivec3 pads = ivec3(${t}, ${n}, ${s});

      void main() {
        ivec5 coords = getOutputCoords();
        int batch = coords.x;
        int d2 = coords.u;

        ivec3 xFRCCorner = ivec3(coords.y, coords.z, coords.w) * strides - pads;
        int xFCorner = xFRCCorner.x;
        int xRCorner = xFRCCorner.y;
        int xCCorner = xFRCCorner.z;

        // Convolve x(?, ?, ?, d1) with w(:, :, :, d1, d2) to get
        // y(yF, yR, yC, d2). ? = to be determined. : = across all
        // values in that axis.
        float dotProd = 0.0;
        for (int wF = 0; wF < ${c}; wF++) {
          int xF = xFCorner + wF * ${o};

          if (xF < 0 || xF >= ${e.inDepth}) {
            continue;
          }

          for (int wR = 0; wR < ${p}; wR++) {
            int xR = xRCorner + wR * ${u};

            if (xR < 0 || xR >= ${e.inHeight}) {
              continue;
            }

            for (int wC = 0; wC < ${d}; wC++) {
              int xC = xCCorner + wC * ${l};

              if (xC < 0 || xC >= ${e.inWidth}) {
                continue;
              }

              for (int d1 = 0; d1 < ${h}; d1 += 4) {
                vec4 xValues = vec4(
                  getX(batch, xF, xR, xC, d1),
                  getX(batch, xF, xR, xC, d1 + 1),
                  getX(batch, xF, xR, xC, d1 + 2),
                  getX(batch, xF, xR, xC, d1 + 3)
                );
                vec4 wValues = vec4(
                  getW(wF, wR, wC, d1, d2),
                  getW(wF, wR, wC, d1 + 1, d2),
                  getW(wF, wR, wC, d1 + 2, d2),
                  getW(wF, wR, wC, d1 + 3, d2)
                );

                dotProd += dot(xValues, wValues);
              }

              if (${f === 1}) {
                dotProd +=
                  getX(batch, xF, xR, xC, ${h}) *
                  getW(wF, wR, wC, ${h}, d2);
              } else if (${f === 2}) {
                vec2 xValues = vec2(
                  getX(batch, xF, xR, xC, ${h}),
                  getX(batch, xF, xR, xC, ${h} + 1)
                );
                vec2 wValues = vec2(
                  getW(wF, wR, wC, ${h}, d2),
                  getW(wF, wR, wC, ${h} + 1, d2)
                );
                dotProd += dot(xValues, wValues);
              } else if (${f === 3}) {
                vec3 xValues = vec3(
                  getX(batch, xF, xR, xC, ${h}),
                  getX(batch, xF, xR, xC, ${h} + 1),
                  getX(batch, xF, xR, xC, ${h} + 2)
                );
                vec3 wValues = vec3(
                  getW(wF, wR, wC, ${h}, d2),
                  getW(wF, wR, wC, ${h} + 1, d2),
                  getW(wF, wR, wC, ${h} + 2, d2)
                );
                dotProd += dot(xValues, wValues);
              }
            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
var f9 = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true, this.customUniforms = [{ name: "inputShape", type: "ivec3" }, { name: "pad", type: "ivec2" }, { name: "stride", type: "ivec2" }, { name: "dilation", type: "ivec2" }, { name: "inChannels", type: "int" }, { name: "itemsPerBlockRow", type: "int" }, { name: "outWidth", type: "int" }], this.outputShape = e, this.enableShapeUniforms = Sn(this.outputShape.length);
    let { dataFormat: n } = t, s = fn(), r = n === "channelsLast", a = r ? 0 : 1, i = r ? 1 : 2, o = this.enableShapeUniforms ? "if(blockIndex < outShape[1] && pos < outShape[0]) {" : `if(blockIndex < ${e[1]} && pos < ${e[0]}) {`, u = "";
    for (let l = 0; l <= 1; l++)
      for (let c = 0; c <= 1; c++)
        u += `
          blockIndex = rc.y + ${c};
          pos = rc.x + ${l};

          ${o}
            offsetY = int(blockIndex / outWidth) * stride[0] - pad[0];
            d0 = offsetY + dilation[0] * (pos / itemsPerBlockRow);

            if(d0 < inputShape[${a}] && d0 >= 0) {
              // Use custom imod instead mod. On Intel GPU, mod may generate
              // unexpected value.
              // https://github.com/tensorflow/tfjs/issues/5447
              offsetX = imod(blockIndex, outWidth) * stride[1] - pad[1];
              d1 = offsetX + dilation[1] * (imod(pos, itemsPerBlockRow) /
                  inChannels);

              if(d1 < inputShape[${i}] && d1 >= 0) {

                ch = imod(pos, inChannels);

                if (${r}) {
                  innerDims = vec2(d1, ch);
                  result[${l * 2 + c}] = getChannel(
                    getA(d0, int(innerDims.x),
                    int(innerDims.y)), innerDims);
                } else {
                  innerDims = vec2(d0, d1);
                  result[${l * 2 + c}] = getChannel(
                    getA(ch, int(innerDims.x),
                    int(innerDims.y)), innerDims);
                }
              }
            }
          }
        `;
    this.userCode = `
      void main() {
        ivec2 rc = getOutputCoords();

        vec4 result = vec4(0);

        int blockIndex, pos, offsetY, d0, offsetX, d1, ch;
        vec2 innerDims;

        ${u}

        ${s.output} = result;
      }
    `;
  }
};
function u2({ x: e, filter: t, convInfo: n, backend: s, bias: r = null, preluActivationWeights: a = null, leakyreluAlpha: i = 0, activation: o = null }) {
  let u = e.shape, l = s.texData.get(e.dataId), c = n.inChannels, p = u[0] * u[1] * u[2], d = n.outChannels, h = n.dataFormat === "channelsLast", f = false, m = false, g, b = [];
  if (a != null && !h && a.shape.length === 3) {
    let x = _t({ inputs: { x: a }, backend: s, attrs: { perm: [1, 2, 0] } });
    b.push(x), a = x;
  }
  if (!((p === 1 || d === 1) && c > t2) && l.isPacked && h && l.texture != null && u[2] % 2 !== 0 && w.arraysEqual(l.shape.slice(-3), u.slice(-3))) {
    let x = u[0] * u[1] * (u[2] + 1), k = { dataId: e.dataId, shape: [1, x, n.inChannels], dtype: e.dtype }, I = l.shape;
    l.shape = l.shape.slice(), l.shape[l.shape.length - 2]++, w.assert(rl(l.shape, k.shape), () => `packed reshape ${l.shape} to ${k.shape} isn't free`);
    let $ = he({ inputs: { x: t }, backend: s, attrs: { shape: [1, n.inChannels, n.outChannels] } });
    b.push($);
    let R = Gd({ a: k, b: $, backend: s, transposeA: f, transposeB: m, bias: r, activation: o, preluActivationWeights: a, leakyreluAlpha: i }), E = s.texData.get(R.dataId);
    w.assert(E.isPacked, () => "batchMatMul result is expected to be packed"), l.shape = I, E.shape = n.outShape, g = Rn({ inputs: { x: R }, backend: s }), g.shape = n.outShape, b.push(R);
  } else {
    let x = h ? e : _t({ inputs: { x: e }, backend: s, attrs: { perm: [0, 2, 3, 1] } }), k = x.shape, I = k[0] * k[1] * k[2], $ = he({ inputs: { x }, backend: s, attrs: { shape: [1, I, n.inChannels] } }), R = he({ inputs: { x: t }, backend: s, attrs: { shape: [1, n.inChannels, n.outChannels] } }), E = Gd({ a: $, b: R, transposeA: f, transposeB: m, backend: s, bias: r, activation: o, preluActivationWeights: a, leakyreluAlpha: i }), P = [n.batchSize, n.outHeight, n.outWidth, n.outChannels], A = he({ inputs: { x: E }, backend: s, attrs: { shape: P } });
    g = h ? A : _t({ inputs: { x: A }, backend: s, attrs: { perm: [0, 3, 1, 2] } }), h || (b.push(x), b.push(A)), b.push($), b.push(R), b.push(E);
  }
  for (let x of b)
    s.disposeIntermediateTensorInfo(x);
  return g;
}
function l2({ x: e, filter: t, convInfo: n, backend: s, bias: r = null, preluActivationWeights: a = null, leakyreluAlpha: i = 0, activation: o = null }) {
  let { filterWidth: u, filterHeight: l, inChannels: c, outWidth: p, outHeight: d, dataFormat: h } = n, f = h === "channelsLast", m = u * l * c, g = d * p, b = [m, g], y = true, v = false, x = [];
  if (a != null && !f && a.shape.length === 3) {
    let J = _t({ inputs: { x: a }, backend: s, attrs: { perm: [1, 2, 0] } });
    x.push(J), a = J;
  }
  let k = he({ inputs: { x: e }, backend: s, attrs: { shape: e.shape.slice(1) } }), I = he({ inputs: { x: t }, backend: s, attrs: { shape: [1, m, w.sizeFromShape(t.shape) / m] } });
  x.push(k), x.push(I);
  let $ = new f9(b, n), R = [k.shape, [n.padInfo.top, n.padInfo.left], [n.strideHeight, n.strideWidth], [n.dilationHeight, n.dilationWidth], [n.inChannels], [n.filterWidth * n.inChannels], [n.outWidth]], E = s.runWebGLProgram($, [k], "float32", R), P = he({ inputs: { x: E }, backend: s, attrs: { shape: [1, b[0], b[1]] } });
  x.push(E), x.push(P);
  let A = r != null, O = a != null, T = o === "leakyrelu", z = o ? th(o, true) : null, W = new e2(P.shape, I.shape, [1, g, n.outChannels], y, v, A, z, O, T), q = [P, I];
  if (r && q.push(r), O && q.push(a), T) {
    let J = s.makeTensorInfo([], "float32", w.createScalarValue(i, "float32"));
    q.push(J), x.push(J);
  }
  let X = s.runWebGLProgram(W, q, "float32"), Y = [1, d, p, n.outChannels], Z = he({ inputs: { x: X }, backend: s, attrs: { shape: Y } }), te = f ? Z : _t({ inputs: { x: Z }, backend: s, attrs: { perm: [0, 3, 1, 2] } });
  f || x.push(Z), x.push(X);
  for (let J of x)
    s.disposeIntermediateTensorInfo(J);
  return te;
}
function m9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dataFormat: u, dilations: l, dimRoundingMode: c } = s, p = C.convertConv2DDataFormat(u), d = C.computeConv2DInfo(r.shape, a.shape, i, l, o, c, false, p), h;
  if (d.filterHeight === 1 && d.filterWidth === 1 && d.dilationHeight === 1 && d.dilationWidth === 1 && d.strideHeight === 1 && d.strideWidth === 1 && (d.padInfo.type === "SAME" || d.padInfo.type === "VALID"))
    h = u2({ x: r, filter: a, convInfo: d, backend: n });
  else if (K().getBool("WEBGL_CONV_IM2COL") && r.shape[0] === 1)
    h = l2({ x: r, filter: a, convInfo: d, backend: n });
  else {
    let m = new o2(d);
    h = n.runWebGLProgram(m, [r, a], "float32");
  }
  let f = he({ inputs: { x: h }, backend: n, attrs: { shape: d.outShape } });
  return n.disposeIntermediateTensorInfo(h), f;
}
var g9 = { kernelName: _a, backendName: "webgl", kernelFunc: m9 };
var b9 = class {
  constructor(e) {
    this.variableNames = ["x", "dy"], this.outputShape = e.filterShape;
    let t = e.strideHeight, n = e.strideWidth, s = e.padInfo.top, r = e.padInfo.left, a = e.dataFormat === "channelsLast";
    this.userCode = `
      void main() {
        ivec4 coords = getOutputCoords();
        int wR = coords.x;
        int wC = coords.y;
        int d1 = coords.z;
        int d2 = coords.w;

        // Convolve x(?, ?, d1) with dy(:, :, d2) to get dw(wR, wC, d1, d2).
        // ? = to be determined. : = across all values in that axis.
        float dotProd = 0.0;

        for (int b = 0; b < ${e.batchSize}; b++) {
          for (int yR = 0; yR < ${e.outHeight}; yR++) {
            int xR = wR + yR * ${t} - ${s};

            if (xR < 0 || xR >= ${e.inHeight}) {
              continue;
            }

            for (int yC = 0; yC < ${e.outWidth}; yC++) {
              int xC = wC + yC * ${n} - ${r};

              if (xC < 0 || xC >= ${e.inWidth}) {
                continue;
              }

              if (${a}) {
                float dyValue = getDy(b, yR, yC, d2);
                float xValue = getX(b, xR, xC, d1);
                dotProd += (xValue * dyValue);
              } else {
                float dyValue = getDy(b, d2, yR, yC);
                float xValue = getX(b, d1, xR, xC);
                dotProd += (xValue * dyValue);
              }

            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
var y9 = class {
  constructor(e) {
    this.variableNames = ["dy", "W"], this.outputShape = e.inShape;
    let t = e.filterHeight, n = e.filterWidth, s = e.strideHeight, r = e.strideWidth, a = e.dataFormat === "channelsLast", i = t - 1 - e.padInfo.top, o = n - 1 - e.padInfo.left, u = a ? 1 : 2, l = a ? 2 : 3, c = a ? 3 : 1;
    this.userCode = `
      const ivec2 pads = ivec2(${i}, ${o});

      void main() {
        ivec4 coords = getOutputCoords();
        int batch = coords[0];
        int d1 = coords[${c}];

        ivec2 dyCorner = ivec2(coords[${u}], coords[${l}]) - pads;
        int dyRCorner = dyCorner.x;
        int dyCCorner = dyCorner.y;

        // Convolve dy(?, ?, d2) with w(:, :, d1, d2) to compute dx(xR, xC, d1).
        // ? = to be determined. : = across all values in that axis.
        float dotProd = 0.0;
        for (int wR = 0; wR < ${t}; wR++) {
          float dyR = float(dyRCorner + wR) / ${s}.0;

          if (dyR < 0.0 || dyR >= ${e.outHeight}.0 || fract(dyR) > 0.0) {
            continue;
          }
          int idyR = int(dyR);

          int wRPerm = ${t} - 1 - wR;

          for (int wC = 0; wC < ${n}; wC++) {
            float dyC = float(dyCCorner + wC) / ${r}.0;

            if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
                fract(dyC) > 0.0) {
              continue;
            }
            int idyC = int(dyC);

            int wCPerm = ${n} - 1 - wC;

            for (int d2 = 0; d2 < ${e.outChannels}; d2++) {

              if (${a}) {
                float xValue = getDy(batch, idyR, idyC, d2);
                float wValue = getW(wRPerm, wCPerm, d1, d2);
                dotProd += xValue * wValue;
              } else {
                float xValue = getDy(batch, d2, idyR, idyC);
                float wValue = getW(wRPerm, wCPerm, d1, d2);
                dotProd += xValue * wValue;
              }

            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
var v9 = class {
  constructor(e) {
    this.variableNames = ["x", "dy"], this.outputShape = e.filterShape;
    let t = e.strideDepth, n = e.strideHeight, s = e.strideWidth, r = e.padInfo.front, a = e.padInfo.top, i = e.padInfo.left;
    this.userCode = `
      void main() {
        ivec5 coords = getOutputCoords();
        int wF = coords.x;
        int wR = coords.y;
        int wC = coords.z;
        int d1 = coords.w;
        int d2 = coords.u;

        float dotProd = 0.0;

        for (int b = 0; b < ${e.batchSize}; b++) {
          for (int yF = 0; yF < ${e.outDepth}; yF++) {
            int xF = wF + yF * ${t} - ${r};

            if (xF < 0 || xF >= ${e.inDepth}) {
              continue;
            }

            for (int yR = 0; yR < ${e.outHeight}; yR++) {
              int xR = wR + yR * ${n} - ${a};

              if (xR < 0 || xR >= ${e.inHeight}) {
                continue;
              }

              for (int yC = 0; yC < ${e.outWidth}; yC++) {
                int xC = wC + yC * ${s} - ${i};

                if (xC < 0 || xC >= ${e.inWidth}) {
                  continue;
                }

                float dyValue = getDy(b, yF, yR, yC, d2);
                float xValue = getX(b, xF, xR, xC, d1);
                dotProd += (xValue * dyValue);
              }
            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
var x9 = class {
  constructor(e) {
    this.variableNames = ["dy", "W"], this.outputShape = e.inShape;
    let t = e.filterDepth, n = e.filterHeight, s = e.filterWidth, r = e.strideDepth, a = e.strideHeight, i = e.strideWidth, o = t - 1 - e.padInfo.front, u = n - 1 - e.padInfo.top, l = s - 1 - e.padInfo.left;
    this.userCode = `
      const ivec3 pads = ivec3(${o}, ${u}, ${l});

      void main() {
        ivec5 coords = getOutputCoords();
        int batch = coords.x;
        int d1 = coords.u;


        ivec3 dyCorner = ivec3(coords.y, coords.z, coords.w) - pads;
        int dyFCorner = dyCorner.x;
        int dyRCorner = dyCorner.y;
        int dyCCorner = dyCorner.z;

        float dotProd = 0.0;
        for (int wF = 0; wF < ${t}; wF++) {
          float dyF = float(dyFCorner + wF) / ${r}.0;

          if (dyF < 0.0 || dyF >= ${e.outDepth}.0 || fract(dyF) > 0.0) {
            continue;
          }
          int idyF = int(dyF);

          int wFPerm = ${t} - 1 - wF;

          for (int wR = 0; wR < ${n}; wR++) {
            float dyR = float(dyRCorner + wR) / ${a}.0;

            if (dyR < 0.0 || dyR >= ${e.outHeight}.0 ||
              fract(dyR) > 0.0) {
              continue;
            }
            int idyR = int(dyR);

            int wRPerm = ${n} - 1 - wR;

            for (int wC = 0; wC < ${s}; wC++) {
              float dyC = float(dyCCorner + wC) / ${i}.0;

              if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
                  fract(dyC) > 0.0) {
                continue;
              }
              int idyC = int(dyC);

              int wCPerm = ${s} - 1 - wC;

              for (int d2 = 0; d2 < ${e.outChannels}; d2++) {
                float xValue = getDy(batch, idyF, idyR, idyC, d2);
                float wValue = getW(wFPerm, wRPerm, wCPerm, d1, d2);
                dotProd += xValue * wValue;
              }
            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
function w9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, dy: a } = t, { strides: i, pad: o, dataFormat: u, dimRoundingMode: l, filterShape: c } = s, p = C.convertConv2DDataFormat(u), d = C.computeConv2DInfo(r.shape, c, i, 1, o, l, false, p), h = new b9(d);
  return n.runWebGLProgram(h, [r, a], "float32");
}
var k9 = { kernelName: gg, backendName: "webgl", kernelFunc: w9 };
function S9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, filter: a } = t, { inputShape: i, strides: o, pad: u, dataFormat: l, dimRoundingMode: c } = s, p = C.convertConv2DDataFormat(l), d = C.computeConv2DInfo(i, a.shape, o, 1, u, c, false, p), h = new y9(d);
  return n.runWebGLProgram(h, [r, a], "float32");
}
var I9 = { kernelName: Aa, backendName: "webgl", kernelFunc: S9 };
function C9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dilations: u } = s, l = C.computeConv3DInfo(r.shape, a.shape, i, u, o), c = new h9(l);
  return n.runWebGLProgram(c, [r, a], "float32");
}
var N9 = { kernelName: np, backendName: "webgl", kernelFunc: C9 };
function T9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, dy: a } = t, { strides: i, pad: o, filterShape: u } = s, l = C.computeConv3DInfo(r.shape, u, i, 1, o), c = new v9(l);
  return n.runWebGLProgram(c, [r, a], "float32");
}
var $9 = { kernelName: bg, backendName: "webgl", kernelFunc: T9 };
function _9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, filter: a } = t, { pad: i, strides: o, inputShape: u } = s, l = C.computeConv3DInfo(u, a.shape, o, 1, i), c = new x9(l);
  return n.runWebGLProgram(c, [r, a], "float32");
}
var A9 = { kernelName: yg, backendName: "webgl", kernelFunc: _9 };
var E9 = du + `
  return cos(x);
`;
var R9 = Ke({ opSnippet: E9 });
var D9 = { kernelName: Ea, backendName: "webgl", kernelFunc: R9 };
var F9 = `
  float e2x = exp(-x);
  return (e2x + 1.0 / e2x) / 2.0;
`;
var O9 = Ke({ opSnippet: F9 });
var P9 = { kernelName: Ra, backendName: "webgl", kernelFunc: O9 };
var z9 = class {
  constructor(e, t, n, s, r) {
    this.variableNames = ["Image", "Boxes", "BoxInd"], this.outputShape = [];
    let [a, i, o, u] = e, [l] = t, [c, p] = n;
    this.outputShape = [l, c, p, u];
    let d = s === "bilinear" ? 1 : 0, [h, f] = [`${i - 1}.0`, `${o - 1}.0`], [m, g, b] = c > 1 ? [`${(i - 1) / (c - 1)}`, "(y2-y1) * height_ratio", `y1*${h} + float(y)*(height_scale)`] : ["0.0", "0.0", `0.5 * (y1+y2) * ${h}`], [y, v, x] = p > 1 ? [`${(o - 1) / (p - 1)}`, "(x2-x1) * width_ratio", `x1*${f} + float(x)*(width_scale)`] : ["0.0", "0.0", `0.5 * (x1+x2) * ${f}`];
    this.userCode = `
      const float height_ratio = float(${m});
      const float width_ratio = float(${y});
      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int y = coords[1];
        int x = coords[2];
        int d = coords[3];

        // get box vals
        float y1 = getBoxes(b,0);
        float x1 = getBoxes(b,1);
        float y2 = getBoxes(b,2);
        float x2 = getBoxes(b,3);

        // get image in batch index
        int bInd = round(getBoxInd(b));
        if(bInd < 0 || bInd >= ${a}) {
          return;
        }

        float height_scale = ${g};
        float width_scale = ${v};

        float in_y = ${b};
        if( in_y < 0.0 || in_y > ${h} ) {
          setOutput(float(${r}));
          return;
        }
        float in_x = ${x};
        if( in_x < 0.0 || in_x > ${f} ) {
          setOutput(float(${r}));
          return;
        }

        vec2 sourceFracIndexCR = vec2(in_x,in_y);
        if(${d} == 1) {
          // Compute the four integer indices.
          ivec2 sourceFloorCR = ivec2(sourceFracIndexCR);
          ivec2 sourceCeilCR = ivec2(ceil(sourceFracIndexCR));

          float topLeft = getImage(b, sourceFloorCR.y, sourceFloorCR.x, d);
          float bottomLeft = getImage(b, sourceCeilCR.y, sourceFloorCR.x, d);
          float topRight = getImage(b, sourceFloorCR.y, sourceCeilCR.x, d);
          float bottomRight = getImage(b, sourceCeilCR.y, sourceCeilCR.x, d);

          vec2 fracCR = sourceFracIndexCR - vec2(sourceFloorCR);

          float top = topLeft + (topRight - topLeft) * fracCR.x;
          float bottom = bottomLeft + (bottomRight - bottomLeft) * fracCR.x;
          float newValue = top + (bottom - top) * fracCR.y;
          setOutput(newValue);
        } else {
          // Compute the coordinators of nearest neighbor point.
          ivec2 sourceNearestCR = ivec2(floor(
            sourceFracIndexCR + vec2(0.5,0.5)));
          float newValue = getImage(b, sourceNearestCR.y, sourceNearestCR.x, d);
          setOutput(newValue);
        }
      }
    `;
  }
};
var M9 = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { image: r, boxes: a, boxInd: i } = t, { cropSize: o, method: u, extrapolationValue: l } = s, c = new z9(r.shape, a.shape, o, u, l);
  return n.runWebGLProgram(c, [r, a, i], "float32");
};
var L9 = { kernelName: go, backendName: "webgl", kernelFunc: M9 };
var Sw = class {
  constructor(e, t, n, s) {
    this.variableNames = ["x"], this.customUniforms = [{ name: "index", type: "float" }], this.op = e, this.outputShape = t;
    let r = t.length, a = this.op === "*" ? "1.0" : "0.0", i = n ? a : `getX(${Iw(r, "coords", this.op)})`, o = t[t.length - 1], u = "", l = "";
    n ? (u = s ? `end != ${o - 1}` : "end != 0", l = s ? "end + 1" : "end - 1") : (u = s ? `end + pow2 < ${o}` : "end >= pow2", l = s ? "end + pow2" : "end - pow2"), this.userCode = `
      void main() {
        ${ot(r)} coords = getOutputCoords();
        int end = ${Cw(r, "coords", this.op)};
        float val = ${i};
        int pow2 = int(pow(2.0, index));
        if (${u}) {
          int idx = ${l};
          ${Cw(r, "coords", this.op)} = idx;
          val ${this.op}= getX(${Iw(r, "coords", this.op)});
        }
        setOutput(val);
      }
    `;
  }
};
function Iw(e, t, n) {
  if (e === 1)
    return `${t}`;
  if (e === 2)
    return `${t}.x, ${t}.y`;
  if (e === 3)
    return `${t}.x, ${t}.y, ${t}.z`;
  if (e === 4)
    return `${t}.x, ${t}.y, ${t}.z, ${t}.w`;
  throw Error(`Cumulative ${n} for rank ${e} is not yet supported`);
}
function Cw(e, t, n) {
  if (e === 1)
    return `${t}`;
  if (e === 2)
    return `${t}.y`;
  if (e === 3)
    return `${t}.z`;
  if (e === 4)
    return `${t}.w`;
  throw Error(`Cumulative ${n} for rank ${e} is not yet supported`);
}
function c2(e, t, n, s, r, a) {
  let i = t.shape.length, o = C.getAxesPermutation([s], i), u = t;
  o != null && (u = _t({ inputs: { x: t }, backend: n, attrs: { perm: o } }));
  let l = C.getInnerMostAxes(1, i)[0];
  if (l !== i - 1)
    throw new Error(`WebGL cumprod shader expects an inner-most axis=${t.shape.length - 1} but got axis=${s}`);
  let c = u.shape[l], p = Rn({ inputs: { x: u }, backend: n });
  for (let d = 0; d <= Math.ceil(Math.log2(c)) - 1; d++) {
    let h = new Sw(e, u.shape, false, a), f = [[d]], m = p;
    p = n.runWebGLProgram(h, [p], p.dtype, f), n.disposeIntermediateTensorInfo(m);
  }
  if (r) {
    let d = new Sw(e, u.shape, r, a), h = p;
    p = n.runWebGLProgram(d, [p], p.dtype), n.disposeIntermediateTensorInfo(h);
  }
  if (o != null) {
    let d = C.getUndoAxesPermutation(o), h = _t({ inputs: { x: p }, backend: n, attrs: { perm: d } });
    return n.disposeIntermediateTensorInfo(p), n.disposeIntermediateTensorInfo(u), h;
  }
  return p;
}
function B9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, exclusive: i, reverse: o } = s;
  return c2("*", r, n, a, i, o);
}
var V9 = { kernelName: mo, backendName: "webgl", kernelFunc: B9 };
function W9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, exclusive: i, reverse: o } = s;
  return c2("+", r, n, a, i, o);
}
var U9 = { kernelName: Da, backendName: "webgl", kernelFunc: W9 };
function G9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, weights: a } = t, { size: i, binaryOutput: o } = s;
  if (r.shape.length === 1) {
    let u = n.readSync(r.dataId), l = n.readSync(a.dataId), c = W1(u, l, a.dtype, a.shape, i);
    return n.makeTensorInfo([i], a.dtype, c);
  } else if (r.shape.length === 2) {
    let u = n.bufferSync(r), l = n.bufferSync(a), c = sX(u, l, i, o);
    return n.makeTensorInfo(c.shape, a.dtype, c.values);
  }
  throw new Error(`Error in denseBincount: input must be at most rank 2, but got rank${r.shape.length}.`);
}
var H9 = { kernelName: vg, backendName: "webgl", kernelFunc: G9 };
var q9 = class {
  constructor(e, t, n) {
    this.variableNames = ["x"], this.outputShape = [], this.outputShape = e, this.blockSize = t, this.dataFormat = n, this.userCode = `
    void main() {
      ivec4 coords = getOutputCoords();
      int b = coords[0];
      int h = ${this.getHeightCoordString()};
      int w = ${this.getWidthCoordString()};
      int d = ${this.getDepthCoordString()};

      int in_h = h / ${t};
      int offset_h = imod(h, ${t});
      int in_w = w / ${t};
      int offset_w = imod(w, ${t});
      int offset_d = (offset_h * ${t} + offset_w) *
        ${this.getOutputDepthSize()};
      int in_d = d + offset_d;

      float result = ${this.getInputSamplingString()};
      setOutput(result);
    }
  `;
  }
  getHeightCoordString() {
    return this.dataFormat === "NHWC" ? "coords[1]" : "coords[2]";
  }
  getWidthCoordString() {
    return this.dataFormat === "NHWC" ? "coords[2]" : "coords[3]";
  }
  getDepthCoordString() {
    return this.dataFormat === "NHWC" ? "coords[3]" : "coords[1]";
  }
  getOutputDepthSize() {
    return this.dataFormat === "NHWC" ? this.outputShape[3] : this.outputShape[1];
  }
  getInputSamplingString() {
    return this.dataFormat === "NHWC" ? "getX(b, in_h, in_w, in_d)" : "getX(b, in_d, in_h, in_w)";
  }
};
function j9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockSize: a, dataFormat: i } = s, o = r.shape[0], u = i === "NHWC" ? r.shape[1] : r.shape[2], l = i === "NHWC" ? r.shape[2] : r.shape[3], c = i === "NHWC" ? r.shape[3] : r.shape[1], p = u * a, d = l * a, h = c / (a * a), f = i === "NHWC" ? [o, p, d, h] : [o, h, p, d], m = new q9(f, a, i);
  return n.runWebGLProgram(m, [r], r.dtype);
}
var K9 = { kernelName: bo, backendName: "webgl", kernelFunc: j9 };
var d2 = class {
  constructor(e, t = false, n = null, s = false, r = false) {
    this.variableNames = ["x", "W"], this.customUniforms = [{ name: "pads", type: "ivec2" }, { name: "strides", type: "ivec2" }, { name: "dilations", type: "ivec2" }, { name: "inDims", type: "ivec2" }], this.outputShape = e.outShape, this.enableShapeUniforms = Sn(this.outputShape.length);
    let a = e.filterHeight, i = e.filterWidth, o = e.outChannels / e.inChannels, u = "", l = "";
    n && (s ? u = `float activation(float a) {
          float b = getPreluActivationWeightsAtOutCoords();
          ${n}
        }` : r ? u = `float activation(float a) {
          float b = getLeakyreluAlphaAtOutCoords();
          ${n}
        }` : u = `
          float activation(float x) {
            ${n}
          }
        `, l = "result = activation(result);");
    let c = t ? "result += getBiasAtOutCoords();" : "";
    t && this.variableNames.push("bias"), s && this.variableNames.push("preluActivationWeights"), r && this.variableNames.push("leakyreluAlpha"), this.userCode = `
      ${u}

      void main() {
        ivec4 coords = getOutputCoords();
        int batch = coords.x;
        ivec2 xRCCorner = coords.yz * strides - pads;
        int d2 = coords.w;
        int d1 = d2 / ${o};
        int q = d2 - d1 * ${o};

        int xRCorner = xRCCorner.x;
        int xCCorner = xRCCorner.y;

        // Convolve x(?, ?, d1) with w(:, :, d1, q) to get y(yR, yC, d2).
        // ? = to be determined. : = across all values in that axis.
        float dotProd = 0.0;
        // TO DO(dsmilkov): Flatten the two for loops and vec4 the operations.
        for (int wR = 0; wR < ${a}; wR++) {
          int xR = xRCorner + wR * dilations[0];

          if (xR < 0 || xR >= inDims[0]) {
            continue;
          }

          for (int wC = 0; wC < ${i}; wC++) {
            int xC = xCCorner + wC * dilations[1];

            if (xC < 0 || xC >= inDims[1]) {
              continue;
            }

            float xVal = getX(batch, xR, xC, d1);
            float wVal = getW(wR, wC, d1, q);
            dotProd += xVal * wVal;
          }
        }

        float result = dotProd;
        ${c}
        ${l}
        setOutput(result);
      }
    `;
  }
};
var p2 = class {
  constructor(e, t = false, n = null, s = false, r = false) {
    this.variableNames = ["x", "W"], this.packedInputs = true, this.packedOutput = true, this.customUniforms = [{ name: "pads", type: "ivec2" }, { name: "strides", type: "ivec2" }, { name: "dilations", type: "ivec2" }, { name: "inDims", type: "ivec2" }], this.outputShape = e.outShape, this.enableShapeUniforms = Sn(this.outputShape.length);
    let a = e.outChannels / e.inChannels, i = e.padInfo.left, o = e.strideWidth, u = e.dilationWidth, l = e.filterHeight, c = e.filterWidth, p = c, d = `
      int xR; int xC; int xCOffset;
      vec4 wTexel; vec4 previous; vec4 final;`;
    for (let g = 0; g < c; g++)
      d += `
          vec4 xTexelC${g * 2};
          int xTexelC${g * 2}Ready;
          vec4 xTexelC${g * 2 + 1};
          int xTexelC${g * 2 + 1}Ready;
          vec4 xC${g};`;
    d += `
    for (int r = 0; r < ${l}; r++) {
      `;
    for (let g = 0; g < c; g++)
      d += `
          xTexelC${g * 2} = vec4(0.0);
          xTexelC${g * 2}Ready = 0;
          xTexelC${g * 2 + 1} = vec4(0.0);
          xTexelC${g * 2 + 1}Ready = 0;
          xC${g} = vec4(0.0);`;
    d += `
        xR = xRCorner + r * dilations[0];
        if (xR >=0 && xR < inDims[0]) {
      `;
    for (let g = 0; g < (p + 1) / 2; g++) {
      let b = g * 2;
      if (d += `
          xC = xCCorner + ${b * u};
          `, o === 1) {
        if (b < c && (i % 2 === 1 ? (d += `
                xCOffset = xC + 1;
                if (xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${b}Ready == 0) {
                  xTexelC${b} = getX(batch, xR, xCOffset, d1);

                  // Need to manually clear unused channels in case
                  // we're reading from recycled texture.
                  if (xCOffset + 1 >= inDims[1]) {
                    xTexelC${b}.zw = vec2(0.0);
                  }
                  xTexelC${b}Ready = 1;
                }
              `, u === 1 && b > 0 ? d += `
                xC${b} = vec4(xTexelC${b - 2}.zw, xTexelC${b}.xy);
                ` : d += `
                  xCOffset = xC + 1 - 2;

                  if (xCOffset >= 0 && xCOffset < inDims[1]) {
                    previous = getX(batch, xR, xCOffset, d1);

                    // Need to manually clear unused channels in case
                    // we're reading from recycled texture.
                    if (xCOffset + 1 >= inDims[1]) {
                      previous.zw = vec2(0.0);
                    }

                    xC${b} = vec4(previous.zw, xTexelC${b}.xy);
                  } else {
                    xC${b} = vec4(0.0, 0.0, xTexelC${b}.xy);
                  }
                  `) : d += `
                if (xC >= 0 && xC < inDims[1] && xTexelC${b}Ready == 0) {
                  xTexelC${b} = getX(batch, xR, xC, d1);
                  if (xC + 1 >= inDims[1]) {
                    xTexelC${b}.zw = vec2(0.0);
                  }
                  xTexelC${b}Ready = 1;
                }

                xC${b} = xTexelC${b};
                `, b + 1 < c)) {
          let y = i % 2 === 0 ? w.nearestLargerEven(u) : u;
          u % 2 === 0 && i % 2 === 1 || u % 2 !== 0 && i % 2 !== 1 ? (d += `
                  xCOffset = xC + imod(pads[1], 2) + ${y};

                  if (xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${b + 1}Ready == 0) {
                    xTexelC${b + 1} = getX(batch, xR, xCOffset, d1);

                    // Need to manually clear unused channels in case
                    // we're reading from recycled texture.
                    if (xCOffset + 1 >= inDims[1]) {
                      xTexelC${b + 1}.zw = vec2(0.0);
                    }
                    xTexelC${b + 1}Ready = 1;
                  }
                  `, u > 1 && (d += `
                    xCOffset -= 2;
                    if (xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${b}Ready == 0) {
                      xTexelC${b} = getX(batch, xR, xCOffset, d1);
                      xTexelC${b}Ready = 1;
                    }
                    `), d += `
                  xC${b + 1} = vec4(xTexelC${b}.zw, xTexelC${b + 1}.xy);
                  `) : y === 1 ? d += `
                    xC${b + 1} = xTexelC${b};
                    ` : d += `
                    xCOffset = xC + ${y};

                    if (xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${b + 1}Ready == 0) {
                      xTexelC${b + 1} = getX(batch, xR, xCOffset, d1);
                      if (xCOffset + 1 >= inDims[1]) {
                        xTexelC${b + 1}.zw = vec2(0.0);
                      }
                      xTexelC${b + 1}Ready = 1;
                    }

                    xC${b + 1} = xTexelC${b + 1};
                    `;
        }
      } else
        b < c && (i % 2 === 1 ? (d += `
                xCOffset = xC + 1 - strides[1];
                if(xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${b}Ready == 0) {
                  xTexelC${b} = getX(batch, xR, xCOffset, d1);
                  // Need to manually clear unused channels in case
                  // we're reading from recycled texture.
                  if (xCOffset + 1 >= inDims[1]) {
                    xTexelC${b}.zw = vec2(0.0);
                  }
                  xTexelC${b}Ready = 1;
                }

                if(xC + 1 >= 0 && xC + 1 < inDims[1] && xTexelC${b + 1}Ready == 0) {
                  xTexelC${b + 1} = getX(batch, xR, xC + 1, d1);
                  // Need to manually clear unused channels in case
                  // we're reading from recycled texture.
                  if (xC + 2 >= inDims[1]) {
                    xTexelC${b + 1}.zw = vec2(0.0);
                  }
                  xTexelC${b + 1}Ready = 1;
                }

                xC${b} = vec4(xTexelC${b}.zw, xTexelC${b + 1}.zw);
              `, b + 1 < c && (d += `
                  final = vec4(0.0);
                  xCOffset = xC + 1 + strides[1];
                  if(xCOffset >= 0 && xCOffset < inDims[1]) {
                    final = getX(batch, xR, xCOffset, d1);
                  }
                  xC${b + 1} = vec4(xTexelC${b + 1}.xy, final.xy);
                `)) : (d += `
                if(xC >= 0 && xC < inDims[1] && xTexelC${b}Ready == 0) {
                  xTexelC${b} = getX(batch, xR, xC, d1);
                  if (xC + 1 >= inDims[1]) {
                    xTexelC${b}.zw = vec2(0.0);
                  }
                  xTexelC${b}Ready = 1;
                }

                xCOffset = xC + strides[1];
                if(xCOffset >= 0 && xCOffset < inDims[1] && xTexelC${b + 1}Ready == 0) {
                  xTexelC${b + 1} = getX(batch, xR, xCOffset, d1);
                  if (xCOffset + 1 >= inDims[1]) {
                    xTexelC${b + 1}.zw = vec2(0.);
                  }
                  xTexelC${b + 1}Ready = 1;
                }

                xC${b} = vec4(
                  xTexelC${b}.xy, xTexelC${b + 1}.xy);
              `, b + 1 < c && (d += `
                  xC${b + 1} = vec4(xTexelC${b}.zw, xTexelC${b + 1}.zw);
                `)));
      b < c && (d += `
            wTexel = getW(r, ${b}, d1, q);
            dotProd += xC${b} * vec4(wTexel.xz, wTexel.xz);
          `, b + 1 < c && (d += `
              wTexel = getW(r, ${b + 1}, d1, q);
              dotProd += xC${b + 1} * vec4(wTexel.xz, wTexel.xz);
            `));
    }
    d += `
    }
  `, d += `
      }
    `;
    let h = "", f = "";
    n && (s ? h = `vec4 activation(vec4 a) {
          vec4 b = getPreluActivationWeightsAtOutCoords();
          ${n}
        }` : r ? h = `vec4 activation(vec4 a) {
          vec4 b = getLeakyreluAlphaAtOutCoords();
          ${n}
        }` : h = `vec4 activation(vec4 x) {
          ${n}
        }`, f = "result = activation(result);");
    let m = t ? "result += getBiasAtOutCoords();" : "";
    t && this.variableNames.push("bias"), s && this.variableNames.push("preluActivationWeights"), r && this.variableNames.push("leakyreluAlpha"), this.userCode = `
      ${h}

      void main() {
        ivec4 coords = getOutputCoords();
        int batch = coords.x;
        ivec2 xRCCorner = coords.yz * strides - pads;
        int d2 = coords.w;
        int d1 = d2 / ${a};
        int q = d2 - d1 * ${a};
        int xRCorner = xRCCorner.x;
        int xCCorner = xRCCorner.y;

        //intialize dotProd with a small epsilon seems to reduce GPU accuracy loss.
        vec4 dotProd = vec4(0.000000000000001);

        ${d}

        vec4 result = dotProd - vec4(0.000000000000001);
        ${m}
        ${f}
        setOutput(result);
      }
    `;
  }
};
function X9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dilations: u, dimRoundingMode: l } = s, c = u;
  c == null && (c = [1, 1]), w.assert(C.eitherStridesOrDilationsAreOne(i, c), () => `Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${i} and dilations '${c}'`);
  let p = C.computeConv2DInfo(r.shape, a.shape, i, c, o, l, true), d;
  K().getBool("WEBGL_PACK_DEPTHWISECONV") && p.strideWidth <= 2 && p.outChannels / p.inChannels === 1 ? d = new p2(p) : d = new d2(p);
  let h = [[p.padInfo.top, p.padInfo.left], [p.strideHeight, p.strideWidth], [p.dilationHeight, p.dilationWidth], [p.inHeight, p.inWidth]];
  return n.runWebGLProgram(d, [r, a], "float32", h);
}
var Y9 = { kernelName: Fa, backendName: "webgl", kernelFunc: X9 };
var Q9 = class {
  constructor(e) {
    this.variableNames = ["x", "dy"], this.outputShape = e.filterShape;
    let t = e.strideHeight, n = e.strideWidth, s = e.padInfo.top, r = e.padInfo.left, a = e.outChannels / e.inChannels;
    this.userCode = `
      void main() {
        ivec4 coords = getOutputCoords();
        int wR = coords.x;
        int wC = coords.y;
        int d1 = coords.z;
        int dm = coords.w;
        int d2 = d1 * ${a} + dm;

        float dotProd = 0.0;

        // TO DO: Vec4 over the batch size
        for (int b = 0; b < ${e.batchSize}; b++) {
          for (int yR = 0; yR < ${e.outHeight}; yR++) {
            int xR = wR + yR * ${t} - ${s};

            if (xR < 0 || xR >= ${e.inHeight}) {
              continue;
            }

            for (int yC = 0; yC < ${e.outWidth}; yC++) {
              int xC = wC + yC * ${n} - ${r};

              if (xC < 0 || xC >= ${e.inWidth}) {
                continue;
              }

              float dyValue = getDy(b, yR, yC, d2);
              float xValue = getX(b, xR, xC, d1);
              dotProd += (xValue * dyValue);
            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
var Z9 = class {
  constructor(e) {
    this.variableNames = ["dy", "W"], this.outputShape = e.inShape;
    let t = e.filterHeight, n = e.filterWidth, s = e.strideHeight, r = e.strideWidth, a = t - 1 - e.padInfo.top, i = n - 1 - e.padInfo.left, o = e.outChannels / e.inChannels;
    this.userCode = `
      const ivec2 pads = ivec2(${a}, ${i});

      void main() {
        ivec4 coords = getOutputCoords();
        int batch = coords[0];
        int d1 = coords[3];
        ivec2 dyCorner = coords.yz - pads;
        int dyRCorner = dyCorner.x;
        int dyCCorner = dyCorner.y;

        float dotProd = 0.0;

        for (int wR = 0; wR < ${t}; wR++) {
          float dyR = float(dyRCorner + wR) / ${s}.0;

          if (dyR < 0.0 || dyR >= ${e.outHeight}.0 || fract(dyR) > 0.0) {
            continue;
          }
          int idyR = int(dyR);

          int wRPerm = ${t} - 1 - wR;

          for (int wC = 0; wC < ${n}; wC++) {
            float dyC = float(dyCCorner + wC) / ${r}.0;

            if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
                fract(dyC) > 0.0) {
              continue;
            }
            int idyC = int(dyC);

            int wCPerm = ${n} - 1 - wC;

            // TO DO: Vec4 over the channelMul
            for (int dm = 0; dm < ${o}; dm++) {
              int d2 = d1 * ${o} + dm;
              float xValue = getDy(batch, idyR, idyC, d2);
              float wValue = getW(wRPerm, wCPerm, d1, dm);
              dotProd += xValue * wValue;
            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
function J9(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, dy: a } = t, { strides: i, dilations: o, pad: u, dimRoundingMode: l, filterShape: c } = s, p = C.computeConv2DInfo(r.shape, c, i, o, u, l, true), d = new Q9(p);
  return n.runWebGLProgram(d, [r, a], "float32");
}
var eQ = { kernelName: xg, backendName: "webgl", kernelFunc: J9 };
function tQ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, filter: a } = t, { strides: i, dilations: o, pad: u, dimRoundingMode: l, inputShape: c } = s, p = C.computeConv2DInfo(c, a.shape, i, o, u, l, true), d = new Z9(p);
  return n.runWebGLProgram(d, [r, a], "float32");
}
var nQ = { kernelName: wg, backendName: "webgl", kernelFunc: tQ };
var sQ = class {
  constructor(e) {
    this.variableNames = ["X"], this.outputShape = [e, e], this.userCode = `
      void main() {
          ivec2 coords = getOutputCoords();
          float val = coords[0] == coords[1] ? getX(coords[0]) : 0.0;
          setOutput(val);
      }
    `;
  }
};
function rQ(e) {
  let { inputs: t, backend: n } = e, { x: s } = t, r = [...s.shape, ...s.shape], a = w.sizeFromShape(s.shape), i = he({ inputs: { x: s }, backend: n, attrs: { shape: [a] } }), o = new sQ(a), u = n.runWebGLProgram(o, [i], i.dtype), l = he({ inputs: { x: u }, backend: n, attrs: { shape: r } });
  return n.disposeIntermediateTensorInfo(i), n.disposeIntermediateTensorInfo(u), l;
}
var aQ = { kernelName: kg, backendName: "webgl", kernelFunc: rQ };
var iQ = class {
  constructor(e) {
    this.variableNames = ["x", "W"], this.outputShape = e.outShape;
    let { inHeight: t, inWidth: n, padInfo: s, strideHeight: r, strideWidth: a, filterHeight: i, filterWidth: o, dilationHeight: u, dilationWidth: l } = e, { top: c, left: p } = s;
    this.userCode = `
      const ivec2 strides = ivec2(${r}, ${a});
      const ivec2 pads = ivec2(${c}, ${p});
      const float neg_infinity = -3.4e38;

      void main() {
        ivec4 coords = getOutputCoords();
        int batch = coords.x;
        int d1 = coords.w;
        ivec2 outTopLeftCorner =
            coords.yz * strides - pads;
        int hBeg = outTopLeftCorner.x;
        int wBeg = outTopLeftCorner.y;

        float curVal = neg_infinity;
        for (int h = 0; h < ${i}; h++) {
          int hIn = hBeg + h * ${u};

          if (hIn >= 0 && hIn < ${t}) {
            for (int w = 0; w < ${o}; w++) {
              int wIn = wBeg + w * ${l};

              if (wIn >= 0 && wIn < ${n}) {
                float xVal = getX(batch, hIn, wIn, d1);
                float wVal = getW(h, w, d1);

                float val = xVal + wVal;
                if (val > curVal) {
                  curVal = val;
                }
              }
            }
          }
        }

        float result = curVal;
        setOutput(result);
      }
    `;
  }
};
function oQ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dilations: u } = s, l = C.computeDilation2DInfo(r.shape, a.shape, i, o, "NHWC", u), c, p = new iQ(l);
  c = n.runWebGLProgram(p, [r, a], "float32");
  let d = he({ inputs: { x: c }, backend: n, attrs: { shape: l.outShape } });
  return n.disposeIntermediateTensorInfo(c), d;
}
var uQ = { kernelName: sp, backendName: "webgl", kernelFunc: oQ };
function lQ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { equation: r } = s, a = t, { allDims: i, summedDims: o, idDims: u } = C.decodeEinsumEquation(r, a.length);
  C.checkEinsumDimSizes(i.length, u, a);
  let { path: l, steps: c } = C.getEinsumComputePath(o, u), p = c.length, d = null, h = i.length, f = [];
  for (let m = 0; m < p; ++m) {
    for (let g of c[m]) {
      let { permutationIndices: b, expandDims: y } = C.getEinsumPermutation(h, u[g]), v;
      C.isIdentityPermutation(b) ? v = a[g] : (v = _t({ inputs: { x: a[g] }, backend: n, attrs: { perm: b } }), f.push(v));
      let x = v.shape.slice();
      for (let k = 0; k < y.length; ++k)
        x.splice(y[k], 0, 1);
      w.arraysEqual(v.shape, x) || (v = he({ inputs: { x: v }, backend: n, attrs: { shape: x } }), f.push(v)), d === null ? d = v : (d = _v({ inputs: { a: v, b: d }, backend: n }), f.push(d));
    }
    m < p - 1 && (l[m] >= 0 && (d = sh({ inputs: { x: d }, backend: n, attrs: { axis: l[m] - (i.length - h), keepDims: false } }), f.push(d)), h--);
  }
  for (let m of f)
    m !== d && n.disposeIntermediateTensorInfo(m);
  return d;
}
var cQ = { kernelName: rp, backendName: "webgl", kernelFunc: lQ };
var dQ = "return (x >= 0.0) ? x : (exp(x) - 1.0);";
var pQ = `
  vec4 result;

  result.r = (x.r >= 0.0) ? x.r : (exp(x.r) - 1.0);
  result.g = (x.g >= 0.0) ? x.g : (exp(x.g) - 1.0);
  result.b = (x.b >= 0.0) ? x.b : (exp(x.b) - 1.0);
  result.a = (x.a >= 0.0) ? x.a : (exp(x.a) - 1.0);

  return result;
`;
var hQ = Ke({ opSnippet: dQ, packedOpSnippet: pQ });
var fQ = { kernelName: Pa, backendName: "webgl", kernelFunc: hQ };
var mQ = "return (b >= 1.0) ? a : a * (b + 1.0);";
var gQ = `
  vec4 bGTEZero = vec4(greaterThanEqual(b, vec4(0.)));
  return (bGTEZero * a) + ((vec4(1.0) - bGTEZero) * (a * (b + vec4(1.0))));
`;
var bQ = (e) => {
  let { inputs: t, backend: n } = e, { dy: s, y: r } = t, a = K().getBool("WEBGL_PACK_BINARY_OPERATIONS") ? new ec(gQ, s.shape, r.shape) : new lo(mQ, s.shape, r.shape);
  return n.runWebGLProgram(a, [s, r], s.dtype);
};
var yQ = { kernelName: Sg, backendName: "webgl", kernelFunc: bQ };
var vQ = `
  return vec4(equal(a, b));
`;
var xQ = "return float(a == b);";
var wQ = jt({ opSnippet: xQ, packedOpSnippet: vQ, dtype: "bool", cpuKernelImpl: iX });
var kQ = { kernelName: yo, backendName: "webgl", kernelFunc: wQ };
var SQ = `
  // Error function is calculated approximately with elementary function.
  // See "Handbook of Mathematical Functions with Formulas,
  // Graphs, and Mathematical Tables", Abramowitz and Stegun.
  float p = ${C.ERF_P};
  float a1 = ${C.ERF_A1};
  float a2 = ${C.ERF_A2};
  float a3 = ${C.ERF_A3};
  float a4 = ${C.ERF_A4};
  float a5 = ${C.ERF_A5};

  float sign = sign(x);
  x = abs(x);
  float t = 1.0 / (1.0 + p * x);
  return sign * (1.0 - (((((a5*t + a4)*t) + a3)*t + a2)*t + a1)*t*exp(-x*x));
`;
var IQ = Ke({ opSnippet: SQ });
var CQ = { kernelName: bl, backendName: "webgl", kernelFunc: IQ };
var NQ = du + `
  return exp(x);
`;
var TQ = `
  vec4 result = exp(x);
  bvec4 isNaN = isnan(x);
  result.r = isNaN.r ? x.r : result.r;
  result.g = isNaN.g ? x.g : result.g;
  result.b = isNaN.b ? x.b : result.b;
  result.a = isNaN.a ? x.a : result.a;

  return result;
`;
var h2 = Ke({ opSnippet: NQ, packedOpSnippet: TQ, cpuKernelImpl: oX, dtype: "float32" });
var $Q = { kernelName: za, backendName: "webgl", kernelFunc: h2 };
function Jm(e) {
  let { inputs: t, attrs: n, backend: s } = e, { dim: r } = n, { input: a } = t, i = a.shape.length, o = a.shape.slice(), u = r;
  return r < 0 && (w.assert(-(i + 1) <= r, () => `Axis must be in the interval [${-(i + 1)}, ${i}]`), u = i + r + 1), o.splice(u, 0, 1), he({ inputs: { x: a }, backend: s, attrs: { shape: o } });
}
var _Q = { kernelName: vo, backendName: "webgl", kernelFunc: Jm };
var Nw = "return exp(x) - 1.0;";
var AQ = Ke({ opSnippet: Nw, packedOpSnippet: Nw, cpuKernelImpl: uX });
var EQ = { kernelName: xo, backendName: "webgl", kernelFunc: AQ };
var Tw = class {
  constructor(e, t, n) {
    this.variableNames = ["real", "imag"];
    let s = t[1];
    this.outputShape = t;
    let r = n ? `2.0 * ${Math.PI}` : `-2.0 * ${Math.PI}`, a = n ? `${s}.0` : "1.0", i;
    if (e === "real")
      i = "return real * expR - imag * expI;";
    else if (e === "imag")
      i = "return real * expI + imag * expR;";
    else
      throw new Error(`FFT component must be either "real" or "imag", got ${e}.`);
    this.userCode = `
      const float exponentMultiplier = ${r};

      float unaryOpComplex(float real, float expR, float imag, float expI) {
        ${i}
      }

      float mulMatDFT(int batch, int index) {
        float indexRatio = float(index) / float(${s});
        float exponentMultiplierTimesIndexRatio =
            exponentMultiplier * indexRatio;

        float result = 0.0;

        for (int i = 0; i < ${s}; i++) {
          // x = (-2|2 * PI / N) * index * i;
          float x = exponentMultiplierTimesIndexRatio * float(i);
          float expR = cos(x);
          float expI = sin(x);
          float real = getReal(batch, i);
          float imag = getImag(batch, i);

          result +=
              unaryOpComplex(real, expR, imag, expI) / ${a};
        }

        return result;
      }

      void main() {
        ivec2 coords = getOutputCoords();
        setOutput(mulMatDFT(coords[0], coords[1]));
      }
    `;
  }
};
function f2(e, t, n) {
  let s = n.texData.get(e.dataId), r = w.sizeFromShape(e.shape), a = e.shape[e.shape.length - 1], i = r / a, o = he({ inputs: { x: e }, backend: n, attrs: { shape: [i, a] } }), u = o.shape, l = new Tw("real", u, t), c = new Tw("imag", u, t), p = [{ dataId: s.complexTensorInfos.real.dataId, dtype: s.complexTensorInfos.real.dtype, shape: u }, { dataId: s.complexTensorInfos.imag.dataId, dtype: s.complexTensorInfos.imag.dtype, shape: u }], d = n.runWebGLProgram(l, p, "float32"), h = n.runWebGLProgram(c, p, "float32"), f = Rr({ inputs: { real: d, imag: h }, backend: n });
  n.disposeIntermediateTensorInfo(d), n.disposeIntermediateTensorInfo(h);
  let m = he({ inputs: { x: f }, backend: n, attrs: { shape: e.shape } });
  return n.disposeIntermediateTensorInfo(o), n.disposeIntermediateTensorInfo(f), m;
}
function RQ(e) {
  let { inputs: t, backend: n } = e, { input: s } = t;
  return f2(s, false, n);
}
var DQ = { kernelName: Ig, backendName: "webgl", kernelFunc: RQ };
var FQ = class {
  constructor(e, t) {
    this.outputShape = [], this.customUniforms = [{ name: "value", type: "float" }], this.variableNames = ["x"], this.outputShape = e, this.userCode = `
      void main() {
        // Input can be obtained from uniform value.
        setOutput(value);
      }
    `;
  }
};
function nc(e) {
  let { backend: t, attrs: n } = e, { shape: s, value: r } = n, { dtype: a } = n;
  if (a = a || w.inferDtype(r), a === "string") {
    let i = w.getArrayFromDType(a, w.sizeFromShape(s));
    return i.fill(r), t.makeTensorInfo(s, a, i);
  } else {
    let i = new FQ(s, r), o = [[r]];
    return t.runWebGLProgram(i, [], a, o);
  }
}
var OQ = { kernelName: yl, backendName: "webgl", kernelFunc: nc };
var PQ = class {
  constructor(e) {
    this.variableNames = ["Image"], this.outputShape = [];
    let t = e[2];
    this.outputShape = e, this.userCode = `
        void main() {
          ivec4 coords = getOutputCoords();
          int x = coords[2];

          int coordX = ${t} - x - 1;
          float outputValue;
          if(coordX >= 0 && coordX < ${t}) {
            outputValue = getImage(coords[0], coords[1], coordX, coords[3]);
          } else {
            outputValue = getImage(coords[0], coords[1], coords[2], coords[3]);
          }
          setOutput(outputValue);
        }
    `;
  }
};
var zQ = { kernelName: wo, backendName: "webgl", kernelFunc: ({ inputs: e, backend: t }) => {
  let { image: n } = e, s = t, r = new PQ(n.shape);
  return s.runWebGLProgram(r, [n], n.dtype);
} };
var $w = "return floor(x);";
var MQ = Ke({ opSnippet: $w, packedOpSnippet: $w, cpuKernelImpl: lX });
var LQ = { kernelName: Ma, backendName: "webgl", kernelFunc: MQ };
var BQ = `
  float s = sign(a) * sign(b);
  int ia = round(a);
  int ib = round(b);
  if (ib != 0) {
    // Windows (D3D) wants guaranteed non-zero int division at compile-time.
    return float(idiv(ia, ib, s));
  } else {
    return NAN;
  }
`;
var VQ = `
  ivec4 ia = round(a);
  ivec4 ib = round(b);
  bvec4 cond = notEqual(ib, ivec4(0));
  ivec4 result = ivec4(0);
  vec4 s = sign(a) * sign(b);

  // Windows (D3D) wants guaranteed non-zero int division at compile-time.
  if (cond[0]) {
    result[0] = idiv(ia[0], ib[0], s[0]);
  }
  if (cond[1]) {
    result[1] = idiv(ia[1], ib[1], s[1]);
  }
  if (cond[2]) {
    result[2] = idiv(ia[2], ib[2], s[2]);
  }
  if (cond[3]) {
    result[3] = idiv(ia[3], ib[3], s[3]);
  }
  return vec4(result);
`;
var WQ = jt({ opSnippet: BQ, packedOpSnippet: VQ, dtype: "int32" });
var UQ = { kernelName: La, backendName: "webgl", kernelFunc: WQ };
var GQ = class {
  constructor(e) {
    this.variableNames = ["A"];
    let t = fn(), [n, s] = e;
    this.outputShape = e, this.userCode = `
      void main() {
        ivec3 coords = getOutputCoords();
        int texR = coords[0];
        int texC = coords[1];
        int depth = coords[2];
        vec2 uv = (vec2(texC, texR) + halfCR) / vec2(${s}.0, ${n}.0);

        vec4 values = ${t.texture2D}(A, uv);
        float value;
        if (depth == 0) {
          value = values.r;
        } else if (depth == 1) {
          value = values.g;
        } else if (depth == 2) {
          value = values.b;
        } else if (depth == 3) {
          value = values.a;
        }

        setOutput(floor(value * 255.0 + 0.5));
      }
    `;
  }
};
var HQ = class {
  constructor(e) {
    this.variableNames = ["A"], this.packedInputs = false, this.packedOutput = true;
    let t = fn(), [n, s] = e;
    this.outputShape = e, this.userCode = `
      void main() {
        ivec3 coords = getOutputCoords();
        int texR = coords[0];
        int texC = coords[1];
        int depth = coords[2];

        vec4 result = vec4(0.);

        for(int row=0; row<=1; row++) {
          for(int col=0; col<=1; col++) {
            texC = coords[1] + row;
            depth = coords[2] + col;

            vec2 uv = (vec2(texC, texR) + halfCR) /
                       vec2(${s}.0, ${n}.0);
            vec4 values = ${t.texture2D}(A, uv);
            float value;
            if (depth == 0) {
              value = values.r;
            } else if (depth == 1) {
              value = values.g;
            } else if (depth == 2) {
              value = values.b;
            } else if (depth == 3) {
              value = values.a;
            }

            result[row * 2 + col] = floor(value * 255.0 + 0.5);
          }
        }

        ${t.output} = result;
      }
    `;
  }
};
var qQ = { kernelName: yd, backendName: "webgl", kernelFunc: jQ };
var Wi;
function jQ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { pixels: r } = t, { numChannels: a } = s, i = typeof HTMLVideoElement != "undefined" && r instanceof HTMLVideoElement, o = typeof HTMLImageElement != "undefined" && r instanceof HTMLImageElement, [u, l] = i ? [r.videoWidth, r.videoHeight] : [r.width, r.height], c = [l, u], p = [l, u, a];
  (o || i) && (Wi == null && (Wi = document.createElement("canvas").getContext("2d")), Wi.canvas.width = u, Wi.canvas.height = l, Wi.drawImage(r, 0, 0, u, l), r = Wi.canvas);
  let d = n.makeTensorInfo(c, "int32");
  n.texData.get(d.dataId).usage = 2, n.gpgpu.uploadPixelDataToTexture(n.getTexture(d.dataId), r);
  let h = K().getBool("WEBGL_PACK") ? new HQ(p) : new GQ(p), f = n.runWebGLProgram(h, [d], "int32");
  return n.disposeData(d.dataId), f;
}
function KQ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a, bias: i, preluActivationWeights: o } = t, { strides: u, pad: l, dataFormat: c, dilations: p, dimRoundingMode: d, activation: h, leakyreluAlpha: f } = s, m = C.convertConv2DDataFormat(c), g = C.computeConv2DInfo(r.shape, a.shape, u, p, l, d, false, m), b, y = [];
  if (g.filterHeight === 1 && g.filterWidth === 1 && g.dilationHeight === 1 && g.dilationWidth === 1 && g.strideHeight === 1 && g.strideWidth === 1 && (g.padInfo.type === "SAME" || g.padInfo.type === "VALID"))
    b = u2({ x: r, filter: a, convInfo: g, backend: n, bias: i, activation: h, preluActivationWeights: o, leakyreluAlpha: f });
  else if (K().getBool("WEBGL_CONV_IM2COL") && r.shape[0] === 1)
    b = l2({ x: r, filter: a, convInfo: g, backend: n, bias: i, activation: h, preluActivationWeights: o, leakyreluAlpha: f });
  else {
    let x = i != null, k = o != null, I = h === "leakyrelu", $ = h ? th(h, false) : null, R = new o2(g, x, $, k, I), E = [r, a];
    if (i && E.push(i), o && E.push(o), I) {
      let P = n.makeTensorInfo([], "float32", w.createScalarValue(f, "float32"));
      E.push(P), y.push(P);
    }
    b = n.runWebGLProgram(R, E, "float32");
  }
  let v = he({ inputs: { x: b }, backend: n, attrs: { shape: g.outShape } });
  return y.push(b), y.forEach((x) => n.disposeIntermediateTensorInfo(x)), v;
}
var XQ = { kernelName: ia, backendName: "webgl", kernelFunc: KQ };
function YQ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a, bias: i, preluActivationWeights: o } = t, { strides: u, pad: l, dilations: c, dimRoundingMode: p, activation: d, leakyreluAlpha: h } = s, f = [], m = c;
  m == null && (m = [1, 1]), w.assert(C.eitherStridesOrDilationsAreOne(u, m), () => `Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${u} and dilations '${m}'`);
  let g = C.computeConv2DInfo(r.shape, a.shape, u, m, l, p, true), b = K().getBool("WEBGL_PACK_DEPTHWISECONV") && g.strideWidth <= 2 && g.outChannels / g.inChannels === 1, y = d ? th(d, b) : null, v = [r, a], x = i != null, k = o != null, I = d === "leakyrelu";
  if (x && v.push(i), k && v.push(o), I) {
    let P = n.makeTensorInfo([], "float32", w.createScalarValue(h, "float32"));
    v.push(P), f.push(P);
  }
  let $;
  b ? $ = new p2(g, x, y, k, I) : $ = new d2(g, x, y, k, I);
  let R = [[g.padInfo.top, g.padInfo.left], [g.strideHeight, g.strideWidth], [g.dilationHeight, g.dilationWidth], [g.inHeight, g.inWidth]], E = n.runWebGLProgram($, v, "float32", R);
  return f.forEach((P) => n.disposeIntermediateTensorInfo(P)), E;
}
var QQ = { kernelName: oa, backendName: "webgl", kernelFunc: YQ };
var ZQ = class {
  constructor(e, t, n) {
    this.sliceDim = e, this.strides = t, this.variableNames = ["x", "indices"], this.outputShape = n;
    let s = ot(t.length), r = ot(n.length), a = this.sliceDim > 1 ? "strides[j]" : "strides";
    this.userCode = `
        ${s} strides = ${s}(${this.strides});
         void main() {
          ${r} coords = getOutputCoords();
          int flattenIndex = 0;
          for (int j = 0; j < ${this.sliceDim}; j++) {
            int index = round(getIndices(coords[0], j));
            flattenIndex += index * ${a};
          }
          setOutput(getX(flattenIndex, coords[1]));
        }
      `;
  }
};
function JQ(e) {
  let { inputs: t, backend: n } = e, { params: s, indices: r } = t, a = r.shape, i = a[a.length - 1], o = w.sizeFromShape(s.shape), [u, l, c, p] = C.prepareAndValidate(s, r), d = he({ inputs: { x: r }, backend: n, attrs: { shape: [l, i] } }), h = he({ inputs: { x: s }, backend: n, attrs: { shape: [w.sizeFromShape(s.shape) / c, c] } });
  if (n.shouldExecuteOnCPU([s, r]) || s.dtype === "string") {
    let b = n.readSync(r.dataId), y = n.bufferSync(s), v = cX(b, y, s.dtype, l, i, c, p, s.shape, o);
    return n.makeTensorInfo(u, s.dtype, v.values);
  }
  let f = new ZQ(i, p, [l, c]), m = n.runWebGLProgram(f, [h, d], h.dtype), g = he({ inputs: { x: m }, backend: n, attrs: { shape: u } });
  return n.disposeIntermediateTensorInfo(d), n.disposeIntermediateTensorInfo(h), n.disposeIntermediateTensorInfo(m), g;
}
var eZ = { kernelName: So, backendName: "webgl", kernelFunc: JQ };
var tZ = class {
  constructor(e, t) {
    this.variableNames = ["A", "indices"], this.outputShape = t, this.rank = t.length;
    let n = ot(this.rank), s = nZ(e, 2);
    this.userCode = `
      void main() {
        ${n} resRC = getOutputCoords();
        int index = int(getIndices(resRC.x, resRC.z));
        float inBounds = (index >= 0) && (index < ${e[2]}) ? 1.0 : 0.0;
        setOutput(inBounds * getA(${s}));
      }
    `;
  }
};
function nZ(e, t) {
  let n = ["resRC.x", "resRC.y", "resRC.z", "resRC.w"], s = [];
  for (let r = 0; r < e.length; r++)
    r === 2 ? s.push("index") : s.push(`${n[r]}`);
  return s.join();
}
function m2(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, indices: a } = t, { axis: i, batchDims: o } = s, u = w.parseAxisParam(i, r.shape)[0];
  if (K().get("DEBUG")) {
    let y = n.readSync(a.dataId), v = r.shape[u];
    for (let x = 0; x < y.length; ++x) {
      let k = y[x];
      w.assert(k <= v - 1 && k >= 0, () => `GatherV2: the index value ${k} is not in [0, ${v - 1}]`);
    }
  }
  let l = C.segment_util.collectGatherOpShapeInfo(r, a, u, o), c = w.sizeFromShape(a.shape), p = [], d = he({ inputs: { x: r }, backend: n, attrs: { shape: [l.batchSize, l.outerSize, l.dimSize, l.sliceSize] } }), h = he({ inputs: { x: a }, backend: n, attrs: { shape: [l.batchSize, c / l.batchSize] } });
  p.push(d), p.push(h);
  let f = [l.batchSize, l.outerSize, c / l.batchSize, l.sliceSize];
  if (n.shouldExecuteOnCPU([r, a]) || r.dtype === "string") {
    let y = n.bufferSync(h), v = n.bufferSync(d), x = dX(v, y, f);
    return p.forEach((k) => n.disposeIntermediateTensorInfo(k)), n.makeTensorInfo(l.outputShape, x.dtype, x.values);
  }
  let m = new tZ(d.shape, f), g = n.runWebGLProgram(m, [d, h], d.dtype);
  p.push(g);
  let b = he({ inputs: { x: g }, backend: n, attrs: { shape: l.outputShape } });
  return p.forEach((y) => n.disposeIntermediateTensorInfo(y)), b;
}
var sZ = { kernelName: ko, backendName: "webgl", kernelFunc: m2 };
var rZ = "return float(a > b);";
var aZ = `
  return vec4(greaterThan(a, b));
`;
var iZ = jt({ opSnippet: rZ, packedOpSnippet: aZ, cpuKernelImpl: pX, dtype: "bool" });
var oZ = { kernelName: Io, backendName: "webgl", kernelFunc: iZ };
var uZ = "return float(a >= b);";
var lZ = `
  return vec4(greaterThanEqual(a, b));
`;
var cZ = jt({ opSnippet: uZ, packedOpSnippet: lZ, dtype: "bool", cpuKernelImpl: hX });
var dZ = { kernelName: Va, backendName: "webgl", kernelFunc: cZ };
function pZ(e) {
  let { inputs: t, backend: n } = e, { input: s } = t;
  return f2(s, true, n);
}
var hZ = { kernelName: Cg, backendName: "webgl", kernelFunc: pZ };
var fZ = "return float(!isnan(x) && !isinf(x));";
var mZ = Ke({ opSnippet: fZ, dtype: "bool" });
var gZ = { kernelName: vl, backendName: "webgl", kernelFunc: mZ };
var bZ = "return float(isinf(x));";
var yZ = Ke({ opSnippet: bZ, dtype: "bool" });
var vZ = { kernelName: xl, backendName: "webgl", kernelFunc: yZ };
var xZ = "return float(isnan(x));";
var wZ = Ke({ opSnippet: xZ, dtype: "bool" });
var kZ = { kernelName: wl, backendName: "webgl", kernelFunc: wZ };
var SZ = "return float(a < b);";
var IZ = `
  return vec4(lessThan(a, b));
`;
var CZ = jt({ opSnippet: SZ, packedOpSnippet: IZ, cpuKernelImpl: fX, dtype: "bool" });
var NZ = { kernelName: Co, backendName: "webgl", kernelFunc: CZ };
var TZ = "return float(a <= b);";
var $Z = `
  return vec4(lessThanEqual(a, b));
`;
var _Z = jt({ opSnippet: TZ, packedOpSnippet: $Z, cpuKernelImpl: mX, dtype: "bool" });
var AZ = { kernelName: No, backendName: "webgl", kernelFunc: _Z };
function EZ(e) {
  let { backend: t, attrs: n } = e, { start: s, stop: r, num: a } = n, i = gX(s, r, a);
  return t.makeTensorInfo([i.length], "float32", i);
}
var RZ = { kernelName: Ng, backendName: "webgl", kernelFunc: EZ };
var DZ = du + `
  return x < 0.0 ? 0./0. : log(x);
`;
var FZ = `
  vec4 result = log(x);
  bvec4 isNaN = isnan(x);
  result.r = isNaN.r ? x.r : (x.r < 0.0 ? 0./0. : result.r);
  result.g = isNaN.g ? x.g : (x.g < 0.0 ? 0./0. : result.g);
  result.b = isNaN.b ? x.b : (x.b < 0.0 ? 0./0. : result.b);
  result.a = isNaN.a ? x.a : (x.a < 0.0 ? 0./0. : result.a);
  return result;
`;
var OZ = Ke({ opSnippet: DZ, packedOpSnippet: FZ, cpuKernelImpl: bX });
var PZ = { kernelName: Ga, backendName: "webgl", kernelFunc: OZ };
var zZ = du + `
  return log(1.0 + x);
`;
var MZ = Ke({ opSnippet: zZ });
var LZ = { kernelName: kl, backendName: "webgl", kernelFunc: MZ };
var BZ = "return float(a >= 1.0 && b >= 1.0);";
var VZ = `
  return vec4(
    vec4(greaterThanEqual(a, vec4(1.0))) *
    vec4(greaterThanEqual(b, vec4(1.0))));
`;
var WZ = jt({ opSnippet: BZ, packedOpSnippet: VZ, dtype: "bool" });
var UZ = { kernelName: To, backendName: "webgl", kernelFunc: WZ };
var GZ = "return float(!(x >= 1.0));";
var HZ = Ke({ opSnippet: GZ });
var qZ = { kernelName: Sl, backendName: "webgl", kernelFunc: HZ };
var jZ = "return float(a >= 1.0 || b >= 1.0);";
var KZ = `
  return min(
    vec4(greaterThanEqual(a, vec4(1.0))) +
    vec4(greaterThanEqual(b, vec4(1.0))),
    vec4(1.0));
`;
var XZ = jt({ opSnippet: jZ, packedOpSnippet: KZ, dtype: "bool" });
var YZ = { kernelName: ip, backendName: "webgl", kernelFunc: XZ };
var QZ = class {
  constructor(e, t, n, s, r) {
    this.variableNames = ["x"], this.outputShape = [];
    let a = t, i = e[3] - 1;
    this.outputShape = e;
    let o, u = `float(${n}) + float(${s}) * sum`;
    r === 0.5 ? o = `inversesqrt(${u})` : r === 1 ? o = `1.0/(${u})` : o = `exp(log(${u}) * float(-${r}));`, this.userCode = `
      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int r = coords[1];
        int c = coords[2];
        int d = coords[3];
        float x = getX(b, r, c, d);
        float sum = 0.0;
        for (int j = -${a}; j <= ${a}; j++) {
          int idx = d + j;
          if (idx >= 0 && idx <=  ${i}) {
            float z = getX(b, r, c, idx);
            sum += z * z;
          }
        }
        float val = x * ${o};
        setOutput(val);
      }
    `;
  }
};
var ZZ = class {
  constructor(e, t, n, s, r) {
    this.variableNames = ["x"], this.outputShape = [], this.packedInputs = true, this.packedOutput = true;
    let a = t, i = e[3] - 1;
    this.outputShape = e;
    let o, u = `float(${n}) + float(${s}) * sum`;
    r === 0.5 ? o = `inversesqrt(${u})` : r === 1 ? o = `1.0/(${u})` : o = `exp(log(${u}) * float(-${r}));`, this.userCode = `
      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords.x;
        int r = coords.y;
        int c = coords.z;
        int d = coords.w;

        bool hasNextCol = d < ${this.outputShape[3]};
        bool hasNextRow = c < ${this.outputShape[2]};

        vec4 sum = vec4(0.);
        vec4 xFragAtOutputCoords = getX(b, r, c, d);

        vec4 xAtOutputCoords = vec4(
          getChannel(xFragAtOutputCoords, vec2(c, d)),
          hasNextCol ?
            getChannel(xFragAtOutputCoords, vec2(c, d + 1)) : 0.0,
          hasNextRow ?
            getChannel(xFragAtOutputCoords , vec2(c + 1, d)) : 0.0,
          (hasNextRow && hasNextCol) ?
            getChannel(xFragAtOutputCoords, vec2(c + 1, d + 1)) : 0.0
        );

        int firstChannel = d - ${a};
        vec2 cache = vec2(0.);
        if(firstChannel >= 0){
          vec4 firstChannelFrag = getX(b, r, c, firstChannel);
          cache.x = getChannel(firstChannelFrag, vec2(c, firstChannel));
            if(hasNextRow){
              cache.y = getChannel(firstChannelFrag, vec2(c + 1, firstChannel));
            }
        }

        ivec2 depth = ivec2(d, d + 1);
        for (int j = - ${a}; j <= ${a}; j++) {
          ivec2 idx = depth + j;
          bvec2 aboveLowerBound = greaterThanEqual(idx, ivec2(0));
          bvec2 belowUpperBound = lessThanEqual(idx, ivec2(${i}));

          bool depthInRange = aboveLowerBound.x && belowUpperBound.x;
          bool depthPlusOneInRange = aboveLowerBound.y && belowUpperBound.y;

          if(depthInRange || depthPlusOneInRange){
            vec4 z = vec4(0.);
            vec4 xFragAtCurrentDepth;
            z.xz = cache.xy;
            if(depthPlusOneInRange && hasNextCol){
              xFragAtCurrentDepth = idx.y != d ?
                getX(b, r, c, idx.y) : xFragAtOutputCoords;
              z.y = getChannel(xFragAtCurrentDepth, vec2(c, idx.y));
              if(hasNextRow){
                z.w = getChannel(xFragAtCurrentDepth, vec2(c + 1, idx.y));
              }
            }
            cache.xy = z.yw;
            sum += z * z;
          }
        }
        vec4 result = xAtOutputCoords * ${o};
        setOutput(result);
      }
    `;
  }
};
var JZ = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { depthRadius: a, bias: i, alpha: o, beta: u } = s, l = K().getBool("WEBGL_PACK_NORMALIZATION") ? new ZZ(r.shape, a, i, o, u) : new QZ(r.shape, a, i, o, u);
  return n.runWebGLProgram(l, [r], r.dtype);
};
var e7 = { kernelName: op, backendName: "webgl", kernelFunc: JZ };
var t7 = class {
  constructor(e, t, n, s, r) {
    this.variableNames = ["inputImage", "outputImage", "dy"], this.outputShape = [], this.outputShape = e, this.depth = e[3], this.depthRadius = t, this.bias = n, this.alpha = s, this.beta = r, this.userCode = `
      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int r = coords[1];
        int c = coords[2];

        float result = 0.0;
        for (int d = 0; d < ${this.depth}; ++d) {
          int depthBegin = int(max(0.0, float(d - ${t})));
          int depthEnd = int(min(float(${this.depth}),
              float(d + ${t} + 1)));

          const int MIN_DEPTH_BEGIN = 0;
          const int MAX_DEPTH_END = ${this.depth};

          float norm = 0.0;
          for (int k = MIN_DEPTH_BEGIN; k < MAX_DEPTH_END; ++k) {
            if (k < depthBegin){
              continue;
            }
            else if (k >= depthBegin && k < depthEnd) {
              norm += getInputImage(b, r, c, k) * getInputImage(b, r, c, k);
            }
            else {
              break;
            }
          }

          norm = float(${s}) * norm + float(${n});

          for(int k = MIN_DEPTH_BEGIN; k < MAX_DEPTH_END; ++k){
            if (k < depthBegin){
              continue;
            }
            else if (k >= depthBegin && k < depthEnd){
              float dyi = -2.0 * float(${s})
                * float(${r})
                * getInputImage(b ,r ,c, k) * getOutputImage(b, r, c, d)
                / norm;
              if (k == d) {
                dyi += pow(norm, -1.0 * ${r});
              }
              if (k == coords[3]) {
                dyi *= getDy(b, r, c, d);
                result += dyi;
              }
            }
            else {
              break;
            }
          }
      }
      setOutput(result);
      }
    `;
  }
};
var n7 = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, y: a, dy: i } = t, { depthRadius: o, bias: u, alpha: l, beta: c } = s, p = new t7(r.shape, o, u, l, c);
  return n.runWebGLProgram(p, [r, a, i], r.dtype);
};
var s7 = { kernelName: Tg, backendName: "webgl", kernelFunc: n7 };
function r7(e, t, n, s) {
  let r = w.sizeFromShape(t), i = w.sizeFromShape(e.shape) / r, o = he({ inputs: { x: e }, attrs: { shape: [i, r] }, backend: s }), u = Si(o, e.dtype, "max", s), l = he({ inputs: { x: u }, attrs: { shape: n }, backend: s });
  return s.disposeIntermediateTensorInfo(o), s.disposeIntermediateTensorInfo(u), l;
}
function g2(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { reductionIndices: a, keepDims: i } = s, o = r.shape.length, u = w.parseAxisParam(a, r.shape), l = u, c = C.getAxesPermutation(l, o), p = c != null, d = n.shouldExecuteOnCPU([r]), h = r;
  if (p) {
    if (d) {
      let v = n.texData.get(h.dataId).values, x = new Array(o);
      for (let $ = 0; $ < x.length; $++)
        x[$] = r.shape[c[$]];
      let k = $v(v, r.shape, r.dtype, c, x);
      h = n.makeTensorInfo(x, r.dtype);
      let I = n.texData.get(h.dataId);
      I.values = k;
    } else
      h = nh(r, c, n);
    l = C.getInnerMostAxes(l.length, o);
  }
  C.assertAxesAreInnerMostDims("max", l, o);
  let [f, m] = C.computeOutAndReduceShapes(h.shape, l), g = f;
  i && (g = C.expandShapeToKeepDim(f, u));
  let b;
  if (d) {
    let v = n.texData.get(h.dataId).values, x = yX(v, w.sizeFromShape(m), g, r.dtype);
    b = n.makeTensorInfo(g, r.dtype);
    let k = n.texData.get(b.dataId);
    k.values = x;
  } else
    b = r7(h, m, g, n);
  return p && n.disposeIntermediateTensorInfo(h), b;
}
var a7 = { kernelName: Ha, backendName: "webgl", kernelFunc: g2 };
var i7 = X1 + `
  return max(a, b);
`;
var o7 = `
  vec4 result = vec4(max(a, b));
  vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
  ` + eh + `
  return result;
`;
var u7 = jt({ opSnippet: i7, packedOpSnippet: o7, cpuKernelImpl: vX });
var l7 = { kernelName: qa, backendName: "webgl", kernelFunc: u7 };
function c7(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t;
  iu(r, "maxPool");
  let { filterSize: a, strides: i, pad: o, dimRoundingMode: u } = s, l = 1;
  w.assert(C.eitherStridesOrDilationsAreOne(i, l), () => `Error in maxPool: Either strides or dilations must be 1. Got strides ${i} and dilations '${l}'`);
  let c = C.computePool2DInfo(r.shape, a, i, l, o, u);
  if (c.filterWidth === 1 && c.filterHeight === 1 && w.arraysEqual(c.inShape, c.outShape))
    return Rn({ inputs: { x: r }, backend: n });
  let p = new al(c, "max", false);
  return n.runWebGLProgram(p, [r], r.dtype);
}
var d7 = { kernelName: ja, backendName: "webgl", kernelFunc: c7 };
function p7(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { filterSize: a, strides: i, pad: o, dataFormat: u, dimRoundingMode: l } = s, c = [1, 1, 1], p = C.computePool3DInfo(r.shape, a, i, c, o, l, u), d = new Av(p, "max", false);
  return n.runWebGLProgram(d, [r], r.dtype);
}
var h7 = { kernelName: up, backendName: "webgl", kernelFunc: p7 };
var f7 = class {
  constructor(e) {
    this.variableNames = ["dy", "maxPos"], this.outputShape = e.inShape;
    let t = e.strideHeight, n = e.strideWidth, s = e.dilationHeight, r = e.effectiveFilterHeight, a = e.effectiveFilterWidth, i = r - 1 - e.padInfo.top, o = a - 1 - e.padInfo.left, u = r * a - 1;
    this.userCode = `
      const ivec2 pads = ivec2(${i}, ${o});

      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int d = coords[3];

        ivec2 dyRCCorner = coords.yz - pads;
        int dyRCorner = dyRCCorner.x;
        int dyCCorner = dyRCCorner.y;

        // Convolve dy(?, ?, d) with pos mask(:, :, d) to get dx(xR, xC, d).
        // ? = to be determined. : = across all values in that axis.
        float dotProd = 0.0;
        for (int wR = 0; wR < ${r};
          wR += ${s}) {
          float dyR = float(dyRCorner + wR) / ${t}.0;

          if (dyR < 0.0 || dyR >= ${e.outHeight}.0 || fract(dyR) > 0.0) {
            continue;
          }
          int idyR = int(dyR);

          for (int wC = 0; wC < ${a}; wC++) {
            float dyC = float(dyCCorner + wC) / ${n}.0;

            if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
                fract(dyC) > 0.0) {
              continue;
            }
            int idyC = int(dyC);

            float dyValue = getDy(b, idyR, idyC, d);
            int maxPosValue = ${u} - int(getMaxPos(b, idyR, idyC, d));

            // Get the current value, check it against the value from the
            // position matrix.
            int curPosValue = wR * ${a} + wC;
            float mask = float(maxPosValue == curPosValue ? 1.0 : 0.0);

            dotProd += dyValue * mask;
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
var m7 = class {
  constructor(e) {
    this.variableNames = ["dy", "maxPos"], this.outputShape = e.inShape;
    let t = e.strideDepth, n = e.strideHeight, s = e.strideWidth, r = e.dilationDepth, a = e.dilationHeight, i = e.dilationWidth, o = e.effectiveFilterDepth, u = e.effectiveFilterHeight, l = e.effectiveFilterWidth, c = o - 1 - e.padInfo.front, p = u - 1 - e.padInfo.top, d = l - 1 - e.padInfo.left, h = o * u * l - 1;
    this.userCode = `
      const ivec3 pads = ivec3(${c}, ${p}, ${d});

      void main() {
        ivec5 coords = getOutputCoords();
        int batch = coords.x;
        int ch = coords.u;

        ivec3 dyCorner = ivec3(coords.y, coords.z, coords.w) - pads;
        int dyDCorner = dyCorner.x;
        int dyRCorner = dyCorner.y;
        int dyCCorner = dyCorner.z;

        // Convolve dy(?, ?, ?, ch) with pos mask(:, :, :, d) to get
        // dx(xD, xR, xC, ch).
        // ? = to be determined. : = across all values in that axis.
        float dotProd = 0.0;

        for (int wD = 0; wD < ${o};
           wD += ${r}) {
          float dyD = float(dyDCorner + wD) / ${t}.0;

          if (dyD < 0.0 || dyD >= ${e.outDepth}.0 || fract(dyD) > 0.0) {
            continue;
          }
          int idyD = int(dyD);

          for (int wR = 0; wR < ${u};
              wR += ${a}) {
            float dyR = float(dyRCorner + wR) / ${n}.0;

            if (dyR < 0.0 || dyR >= ${e.outHeight}.0 ||
                fract(dyR) > 0.0) {
              continue;
            }
            int idyR = int(dyR);

            for (int wC = 0; wC < ${l};
                wC += ${i}) {
              float dyC = float(dyCCorner + wC) / ${s}.0;

              if (dyC < 0.0 || dyC >= ${e.outWidth}.0 ||
                  fract(dyC) > 0.0) {
                continue;
              }
              int idyC = int(dyC);

              float dyValue = getDy(batch, idyD, idyR, idyC, ch);
              int maxPosValue = ${h} -
                  int(getMaxPos(batch, idyD, idyR, idyC, ch));

              // Get the current value, check it against the value from the
              // position matrix.
              int curPosValue =
                  wD * ${u} * ${l} +
                  wR * ${l} + wC;
              float mask = float(maxPosValue == curPosValue ? 1.0 : 0.0);

              dotProd += dyValue * mask;
            }
          }
        }
        setOutput(dotProd);
      }
    `;
  }
};
function g7(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, input: a } = t, i = a, { filterSize: o, strides: u, pad: l, dimRoundingMode: c } = s, p = [1, 1, 1], d = C.computePool3DInfo(i.shape, o, u, p, l, c), h = new Av(d, "max", true), f = n.runWebGLProgram(h, [i], i.dtype), m = new m7(d), g = n.runWebGLProgram(m, [r, f], i.dtype);
  return n.disposeIntermediateTensorInfo(f), g;
}
var b7 = { kernelName: _g, backendName: "webgl", kernelFunc: g7 };
function y7(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, input: a, output: i } = t, o = a;
  iu([a, i], "maxPoolGrad");
  let { filterSize: u, strides: l, pad: c, dimRoundingMode: p } = s, d = C.computePool2DInfo(o.shape, u, l, 1, c, p), h = true, f = new al(d, "max", h), m = n.runWebGLProgram(f, [o], o.dtype), g = new f7(d), b = n.runWebGLProgram(g, [r, m], o.dtype);
  return n.disposeIntermediateTensorInfo(m), b;
}
var v7 = { kernelName: $g, backendName: "webgl", kernelFunc: y7 };
function x7(e, t, n, s) {
  let r = new al(n, "max", false), a = s.runWebGLProgram(r, [e], "float32");
  r = new al(n, "max", true, true, t);
  let i = s.runWebGLProgram(r, [e], "float32");
  return [a, i];
}
var w7 = { kernelName: Ag, backendName: "webgl", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { x: s } = e, { filterSize: r, strides: a, pad: i, includeBatchInIndex: o } = t, u = n;
  w.assert(s.shape.length === 4, () => `Error in maxPool: input must be rank 4 but got rank ${s.shape.length}.`);
  let l = [1, 1];
  w.assert(C.eitherStridesOrDilationsAreOne(a, l), () => `Error in maxPool: Either strides or dilations must be 1. Got strides ${a} and dilations '${l}'`);
  let c = C.computePool2DInfo(s.shape, r, a, l, i), [p, d] = x7(s, o, c, u);
  return [p, d];
} };
function k7(e, t, n, s) {
  let r = w.sizeFromShape(t), i = w.sizeFromShape(e.shape) / r, o = he({ inputs: { x: e }, attrs: { shape: [i, r] }, backend: s }), u = Si(o, "float32", "mean", s), l = he({ inputs: { x: u }, attrs: { shape: n }, backend: s });
  return s.disposeIntermediateTensorInfo(o), s.disposeIntermediateTensorInfo(u), l;
}
var S7 = { kernelName: Ka, backendName: "webgl", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { x: s } = e, { keepDims: r, axis: a } = t, i = n, o = s.shape.length, u = w.parseAxisParam(a, s.shape), l = u, c = C.getAxesPermutation(l, o), p = c != null, d = i.shouldExecuteOnCPU([s]), h = [], f = s;
  if (p) {
    if (d) {
      let x = i.texData.get(f.dataId).values, k = new Array(o);
      for (let R = 0; R < k.length; R++)
        k[R] = s.shape[c[R]];
      let I = $v(x, s.shape, s.dtype, c, k);
      f = i.makeTensorInfo(k, s.dtype);
      let $ = i.texData.get(f.dataId);
      $.values = I;
    } else
      f = nh(s, c, i);
    h.push(f), l = C.getInnerMostAxes(l.length, o);
  }
  C.assertAxesAreInnerMostDims("sum", l, o);
  let [m, g] = C.computeOutAndReduceShapes(f.shape, l), b = m;
  r && (b = C.expandShapeToKeepDim(m, u));
  let y = k7(f, g, b, i);
  for (let v of h)
    i.disposeIntermediateTensorInfo(v);
  return y;
} };
function I7(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s, o = r.shape.length, u = w.parseAxisParam(a, r.shape), l = u, c = C.getAxesPermutation(l, o), p = r;
  c != null && (p = _t({ inputs: { x: r }, backend: n, attrs: { perm: c } }), l = C.getInnerMostAxes(l.length, r.shape.length)), C.assertAxesAreInnerMostDims("min", l, o);
  let [d, h] = C.computeOutAndReduceShapes(p.shape, l), f = w.sizeFromShape(h), m = he({ inputs: { x: p }, backend: n, attrs: { shape: [-1, f] } }), g = Si(m, m.dtype, "min", n), b;
  if (i) {
    let y = C.expandShapeToKeepDim(d, u);
    b = he({ inputs: { x: g }, backend: n, attrs: { shape: y } });
  } else
    b = he({ inputs: { x: g }, backend: n, attrs: { shape: d } });
  return n.disposeIntermediateTensorInfo(m), n.disposeIntermediateTensorInfo(g), c != null && n.disposeIntermediateTensorInfo(p), b;
}
var C7 = { kernelName: Xa, backendName: "webgl", kernelFunc: I7 };
var N7 = X1 + `
  return min(a, b);
`;
var T7 = `
  vec4 result = vec4(min(a, b));
  vec4 isNaN = min(vec4(isnan(a)) + vec4(isnan(b)), vec4(1.0));
  ` + eh + `
  return result;
`;
var $7 = jt({ opSnippet: N7, packedOpSnippet: T7, cpuKernelImpl: xX });
var _7 = { kernelName: Ya, backendName: "webgl", kernelFunc: $7 };
var A7 = class {
  constructor(e, t, n) {
    this.variableNames = ["x"], this.outputShape = t.map((l, c) => l[0] + e[c] + l[1]);
    let s = e.length, r = ot(s), a = t.map((l) => l[0]).join(","), i = t.map((l, c) => l[0] + e[c]).join(","), o = ["coords[0]", "coords[1]", "coords[2]", "coords[3]"].slice(0, s), u = n === "reflect" ? 0 : 1;
    if (s === 1) {
      this.userCode = `
        int start = ${a};
        int end = ${i};

        void main() {
          int outC = getOutputCoords();
          if (outC < start) {
            outC = start * 2 - outC - ${u};
          } else if(outC >= end) {
            outC = (end - 1) * 2 - outC + ${u};
          }
          setOutput(getX(outC - start));
        }
      `;
      return;
    }
    this.userCode = `
      ${r} start = ${r}(${a});
      ${r} end = ${r}(${i});

      void main() {
        ${r} outC = getOutputCoords();
        for (int i = 0; i < ${s}; i++) {
          if (outC[i] < start[i]) {
            outC[i] = start[i] * 2 - outC[i] - ${u};
          } else if(outC[i] >= end[i]) {
            outC[i] = (end[i] - 1) * 2 - outC[i] + ${u};
          }
        }
        ${r} coords = outC - start;
        setOutput(getX(${o}));
      }
    `;
  }
};
var E7 = class {
  constructor(e, t, n) {
    this.variableNames = ["x"], this.packedInputs = true, this.packedOutput = true, this.outputShape = t.map((h, f) => h[0] + e[f] + h[1]);
    let s = e.length, r = ot(s), a = t.map((h) => h[0]).join(","), i = t.map((h, f) => h[0] + e[f]).join(","), o = ln("rc", s), u = ln("source", s), l = `${o[s - 1]} < ${this.outputShape[s - 1]}`, c = s === 1 ? "source" : `vec2(${u.slice(-2).join()})`, p = n === "reflect" ? 0 : 1, d = "";
    if (s === 1) {
      let h = `
        ${r} source = rc;
        if (source < start) {
          source = start * 2 - source - ${p};
        } else if (source >= end) {
          source = (end - 1) * 2 - source + ${p};
        }
        source -= start;
      `;
      d = `
        ${r} rc = outputLoc;
        ${h}
        result[0] = getChannel(getX(${u.join()}), ${c});
        ${o[s - 1]} += 1;
        if(${l}) {
          ${h}
          result[1] = getChannel(getX(${u.join()}), ${c});
        }
      `;
    } else {
      let h = `
        ${r} source = rc;
        ${r} lt = ${r}(lessThan(source, start));
        ${r} gte = ${r}(greaterThanEqual(source, end));
        ${r} orig = 1 - (lt + gte);
        source = orig * source +
                lt * (start * 2 - source - ${p}) +
                gte * ((end - 1) * 2 - source + ${p});
        source -= start;
      `;
      d = `
        ${r} rc = outputLoc;
        ${h}
        result[0] = getChannel(getX(${u.join()}), ${c});
        ${o[s - 1]} += 1;
        if(${l}) {
          ${h}
          result[1] = getChannel(getX(${u.join()}), ${c});
        }
        rc = outputLoc;
        ${o[s - 2]} += 1;
        if(${o[s - 2]} < ${this.outputShape[s - 2]}) {
          ${h}
          result[2] = getChannel(getX(${u.join()}), ${c});
          ${o[s - 1]} += 1;
          if(${l}) {
            ${h}
            result[3] = getChannel(getX(${u.join()}), ${c});
          }
        }
      `;
    }
    this.userCode = `
      const ${r} start = ${r}(${a});
      const ${r} end = ${r}(${i});

      void main() {
        ${r} outputLoc = getOutputCoords();
        vec4 result = vec4(0.);
        ${d}
        setOutput(result);
      }
    `;
  }
};
var R7 = ({ inputs: e, backend: t, attrs: n }) => {
  let { x: s } = e, { paddings: r, mode: a } = n, i = K().getBool("WEBGL_PACK_ARRAY_OPERATIONS") ? new E7(s.shape, r, a) : new A7(s.shape, r, a);
  return t.runWebGLProgram(i, [s], s.dtype);
};
var D7 = { kernelName: Qa, backendName: "webgl", kernelFunc: R7 };
var F7 = `if (b == 0.0) return NAN;
  return mod(a, b);`;
var O7 = `
  vec4 result = mod(a, b);
  vec4 isNaN = vec4(equal(b, vec4(0.0)));
  ` + eh + `
  return result;
`;
var P7 = jt({ opSnippet: F7, packedOpSnippet: O7 });
var z7 = { kernelName: Il, backendName: "webgl", kernelFunc: P7 };
var M7 = class {
  constructor(e, t, n) {
    this.variableNames = ["probs"], this.customUniforms = [{ name: "seed", type: "float" }], this.outputShape = [e, n], this.userCode = `
      void main() {
        ivec2 coords = getOutputCoords();
        int batch = coords[0];

        float r = random(seed);
        float cdf = 0.0;

        for (int i = 0; i < ${t - 1}; i++) {
          cdf += getProbs(batch, i);

          if (r < cdf) {
            setOutput(float(i));
            return;
          }
        }

        // If no other event happened, last event happened.
        setOutput(float(${t - 1}));
      }
    `;
  }
};
var L7 = `
if (a == b) {
  return 1.0;
};
return a / b;`;
var B7 = `
  // vec4 one = vec4(equal(a, b));
  // return one + (vec4(1.0) - one) * a / b;
  vec4 result = a / b;
  if(a.x == b.x) {
    result.x = 1.;
  }
  if(a.y == b.y) {
    result.y = 1.;
  }
  if(a.z == b.z) {
    result.z = 1.;
  }
  if(a.w == b.w) {
    result.w = 1.;
  }

  return result;
`;
var b2 = jt({ opSnippet: L7, packedOpSnippet: B7, checkOutOfBounds: true });
var V7 = { kernelName: Oa, backendName: "webgl", kernelFunc: b2 };
var _w = "return a - b;";
var y2 = jt({ opSnippet: _w, packedOpSnippet: _w, supportsComplex: true, cpuKernelImpl: zX });
var W7 = { kernelName: hi, backendName: "webgl", kernelFunc: y2 };
function v2(e) {
  let { inputs: t, backend: n, attrs: s } = e, { logits: r } = t, { dim: a } = s, i = w.parseAxisParam([a], r.shape), o = g2({ inputs: { x: r }, backend: n, attrs: { reductionIndices: i, keepDims: false } }), u = C.expandShapeToKeepDim(o.shape, i), l = he({ inputs: { x: o }, backend: n, attrs: { shape: u } }), c = y2({ inputs: { a: r, b: l }, backend: n }), p = h2({ inputs: { x: c }, backend: n }), d = sh({ inputs: { x: p }, backend: n, attrs: { axis: i, keepDims: false } }), h = he({ inputs: { x: d }, backend: n, attrs: { shape: u } }), f = b2({ inputs: { a: p, b: h }, backend: n });
  return n.disposeIntermediateTensorInfo(o), n.disposeIntermediateTensorInfo(l), n.disposeIntermediateTensorInfo(c), n.disposeIntermediateTensorInfo(p), n.disposeIntermediateTensorInfo(d), n.disposeIntermediateTensorInfo(h), f;
}
var U7 = { kernelName: di, backendName: "webgl", kernelFunc: v2 };
function G7(e) {
  let { inputs: t, backend: n, attrs: s } = e, { logits: r } = t, { numSamples: a, seed: i, normalized: o } = s, u = o ? r : v2({ inputs: { logits: r }, backend: n, attrs: { dim: r.shape.length - 1 } }), l = u.shape[0], c = u.shape[1], p = new M7(l, c, a), d = [[i]], h = n.runWebGLProgram(p, [u], "int32", d);
  return o || n.disposeIntermediateTensorInfo(u), h;
}
var H7 = { kernelName: Eg, backendName: "webgl", kernelFunc: G7 };
var q7 = ss + `
  return -x;
`;
var j7 = `
  vec4 result = -x;
  bvec4 isNaN = isnan(x);

  result.r = isNaN.r ? x.r : result.r;
  result.g = isNaN.g ? x.g : result.g;
  result.b = isNaN.b ? x.b : result.b;
  result.a = isNaN.a ? x.a : result.a;

  return result;
`;
function K7(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (n.shouldExecuteOnCPU([s])) {
    let a = n.texData.get(s.dataId), [i, o] = kX(a.values, s.shape, s.dtype);
    return n.makeTensorInfo(o, s.dtype, i);
  }
  let r;
  return K().getBool("WEBGL_PACK_UNARY_OPERATIONS") ? r = new Jr(s.shape, j7) : r = new Gs(s.shape, q7), n.runWebGLProgram(r, [s], s.dtype);
}
var X7 = { kernelName: $o, backendName: "webgl", kernelFunc: K7 };
var Y7 = ws.nonMaxSuppressionV3Impl;
function Q7(e) {
  C.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");
  let { inputs: t, backend: n, attrs: s } = e, { boxes: r, scores: a } = t, { maxOutputSize: i, iouThreshold: o, scoreThreshold: u } = s, l = n.readSync(r.dataId), c = n.readSync(a.dataId), { selectedIndices: p } = Y7(l, c, i, o, u);
  return n.makeTensorInfo([p.length], "int32", new Int32Array(p));
}
var Z7 = { kernelName: Ao, backendName: "webgl", kernelFunc: Q7 };
var J7 = ws.nonMaxSuppressionV4Impl;
function eJ(e) {
  C.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");
  let { inputs: t, backend: n, attrs: s } = e, { boxes: r, scores: a } = t, { maxOutputSize: i, iouThreshold: o, scoreThreshold: u, padToMaxOutputSize: l } = s, c = n.readSync(r.dataId), p = n.readSync(a.dataId), { selectedIndices: d, validOutputs: h } = J7(c, p, i, o, u, l);
  return [n.makeTensorInfo([d.length], "int32", new Int32Array(d)), n.makeTensorInfo([], "int32", new Int32Array([h]))];
}
var tJ = { kernelName: Cl, backendName: "webgl", kernelFunc: eJ };
var nJ = ws.nonMaxSuppressionV5Impl;
function sJ(e) {
  C.warn("tf.nonMaxSuppression() in webgl locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");
  let { inputs: t, backend: n, attrs: s } = e, { boxes: r, scores: a } = t, { maxOutputSize: i, iouThreshold: o, scoreThreshold: u, softNmsSigma: l } = s, c = n.readSync(r.dataId), p = n.readSync(a.dataId), d = i, h = o, f = u, m = l, { selectedIndices: g, selectedScores: b } = nJ(c, p, d, h, f, m);
  return [n.makeTensorInfo([g.length], "int32", new Int32Array(g)), n.makeTensorInfo([b.length], "float32", new Float32Array(b))];
}
var rJ = { kernelName: Eo, backendName: "webgl", kernelFunc: sJ };
var aJ = class {
  constructor(e, t, n, s) {
    this.variableNames = ["indices"], this.outputShape = [e, t], this.userCode = `
      void main() {
        ivec2 coords = getOutputCoords();
        int index = round(getIndices(coords.x));
        setOutput(mix(float(${s}), float(${n}),
                      float(index == coords.y)));
      }
    `;
  }
};
var iJ = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { indices: r } = t, { depth: a, onValue: i, offValue: o } = s, u = w.sizeFromShape(r.shape), l = new aJ(u, a, i, o), c = he({ inputs: { x: r }, backend: n, attrs: { shape: [u] } }), p = n.runWebGLProgram(l, [c], r.dtype);
  n.disposeIntermediateTensorInfo(c);
  let d = [...r.shape, a], h = he({ inputs: { x: p }, backend: n, attrs: { shape: d } });
  return n.disposeIntermediateTensorInfo(p), h;
};
var oJ = { kernelName: Do, backendName: "webgl", kernelFunc: iJ };
function Hd(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (s.dtype === "complex64") {
    let r = tc({ inputs: { input: s }, backend: n }), a = Hd({ inputs: { x: r }, backend: n }), i = rh({ inputs: { input: s }, backend: n }), o = Hd({ inputs: { x: i }, backend: n }), u = Rr({ inputs: { real: a, imag: o }, backend: n });
    return n.disposeIntermediateTensorInfo(r), n.disposeIntermediateTensorInfo(a), n.disposeIntermediateTensorInfo(i), n.disposeIntermediateTensorInfo(o), u;
  } else
    return nc({ attrs: { shape: s.shape, dtype: s.dtype, value: s.dtype === "string" ? "" : 0 }, backend: n });
}
var uJ = { kernelName: Xo, backendName: "webgl", kernelFunc: Hd };
function x2(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (s.dtype === "string")
    throw new Error("onesLike is not supported under string dtype");
  if (s.dtype === "complex64") {
    let r = tc({ inputs: { input: s }, backend: n }), a = x2({ inputs: { x: r }, backend: n }), i = rh({ inputs: { input: s }, backend: n }), o = Hd({ inputs: { x: i }, backend: n }), u = Rr({ inputs: { real: a, imag: o }, backend: n });
    return n.disposeIntermediateTensorInfo(r), n.disposeIntermediateTensorInfo(a), n.disposeIntermediateTensorInfo(i), n.disposeIntermediateTensorInfo(o), u;
  } else
    return nc({ attrs: { shape: s.shape, dtype: s.dtype, value: 1 }, backend: n });
}
var lJ = { kernelName: Ro, backendName: "webgl", kernelFunc: x2 };
function cJ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { axis: r } = s;
  if (t.length === 1)
    return Jm({ inputs: { input: t[0] }, backend: n, attrs: { dim: r } });
  let a = t[0].shape, i = t[0].dtype;
  t.forEach((c) => {
    w.assertShapesMatch(a, c.shape, "All tensors passed to stack must have matching shapes"), w.assert(i === c.dtype, () => "All tensors passed to stack must have matching dtypes");
  });
  let o = [], u = t.map((c) => {
    let p = Jm({ inputs: { input: c }, backend: n, attrs: { dim: r } });
    return o.push(p), p;
  }), l = i2({ inputs: u, backend: n, attrs: { axis: r } });
  return o.forEach((c) => n.disposeIntermediateTensorInfo(c)), l;
}
var dJ = { kernelName: Fo, backendName: "webgl", kernelFunc: cJ };
var pJ = class {
  constructor(e, t, n) {
    this.variableNames = ["x"], this.customUniforms = [{ name: "value", type: "float" }], this.outputShape = t.map((u, l) => u[0] + e[l] + u[1]);
    let s = e.length, r = ot(s), a = t.map((u) => u[0]).join(","), i = t.map((u, l) => u[0] + e[l]).join(","), o = ["coords[0]", "coords[1]", "coords[2]", "coords[3]"].slice(0, s);
    if (s === 1) {
      this.userCode = `
        int start = ${a};
        int end = ${i};

        void main() {
          int outC = getOutputCoords();
          if (outC < start || outC >= end) {
            setOutput(value);
          } else {
            setOutput(getX(outC - start));
          }
        }
      `;
      return;
    }
    this.userCode = `
      ${r} start = ${r}(${a});
      ${r} end = ${r}(${i});

      void main() {
        ${r} outC = getOutputCoords();
        if (any(lessThan(outC, start)) || any(greaterThanEqual(outC, end))) {
          setOutput(value);
        } else {
          ${r} coords = outC - start;
          setOutput(getX(${o}));
        }
      }
    `;
  }
};
var hJ = class {
  constructor(e, t, n) {
    this.variableNames = ["x"], this.packedInputs = true, this.packedOutput = true, this.customUniforms = [{ name: "value", type: "float" }], this.outputShape = t.map((f, m) => f[0] + e[m] + f[1]);
    let s = e.length, r = ot(s), a = t.map((f) => f[0]).join(","), i = t.map((f, m) => f[0] + e[m]).join(","), o = ln("rc", s), u = ln("source", s), l = `${o[s - 1]} < ${this.outputShape[s - 1]}`, c = s === 1 ? "source" : `vec2(${u.slice(-2).join()})`, p = [`${r} rc = outputLoc;`, `${o[s - 1]} += 1;
       if(${l}) {
      `, s === 1 ? "" : `}
       rc = outputLoc;
       ${o[s - 2]} += 1;
       if(${o[s - 2]} < ${this.outputShape[s - 2]}) {`, s === 1 ? "" : `  ${o[s - 1]} += 1;
         if(${l}) {`], d = s === 1 ? "rc < start || rc >= end" : "any(lessThan(rc, start)) || any(greaterThanEqual(rc, end))", h = "";
    for (let f = 0, m = s === 1 ? 2 : 4; f < m; f++)
      h += `
        ${p[f]}
        if (${d}) {
          result[${f}] = float(value);
        } else {
          ${r} source = rc - start;
          result[${f}] = getChannel(getX(${u.join()}), ${c});
        }
      `;
    h += s === 1 ? "} " : "}}", this.userCode = `
      const ${r} start = ${r}(${a});
      const ${r} end = ${r}(${i});

      void main() {
        ${r} outputLoc = getOutputCoords();
        vec4 result = vec4(0.);
        ${h}
        setOutput(result);
      }
    `;
  }
};
var w2 = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { paddings: a, constantValue: i } = s;
  if (w.sizeFromShape(r.shape) === 0) {
    let l = a.map((c, p) => c[0] + r.shape[p] + c[1]);
    return nc({ backend: n, attrs: { shape: l, value: i, dtype: r.dtype } });
  }
  let o = K().getBool("WEBGL_PACK_ARRAY_OPERATIONS") ? new hJ(r.shape, a, i) : new pJ(r.shape, a, i), u = [[i]];
  return n.runWebGLProgram(o, [r], r.dtype, u);
};
var fJ = { kernelName: Ja, backendName: "webgl", kernelFunc: w2 };
var mJ = `
  if(a < 0.0 && floor(b) < b){
    return NAN;
  }
  if (b == 0.0) {
    return 1.0;
  }
  return (round(mod(b, 2.0)) != 1) ?
      pow(abs(a), b) : sign(a) * pow(abs(a), b);
`;
var gJ = `
  // isModRound1 has 1 for components with round(mod(b, 2.0)) == 1, 0 otherwise.
  vec4 isModRound1 = vec4(equal(round(mod(b, 2.0)), ivec4(1)));
  vec4 multiplier = sign(a) * isModRound1 + (vec4(1.0) - isModRound1);
  vec4 result = multiplier * pow(abs(a), b);

  // Ensure that a^0 = 1, including 0^0 = 1 as this correspond to TF and JS
  bvec4 isExpZero = equal(b, vec4(0.0));
  result.r = isExpZero.r ? 1.0 : result.r;
  result.g = isExpZero.g ? 1.0 : result.g;
  result.b = isExpZero.b ? 1.0 : result.b;
  result.a = isExpZero.a ? 1.0 : result.a;

  vec4 isNaN = vec4(lessThan(a, vec4(0.0))) * vec4(lessThan(floor(b), b));
  ` + eh + `
  return result;
`;
var bJ = jt({ opSnippet: mJ, packedOpSnippet: gJ });
var yJ = { kernelName: ei, backendName: "webgl", kernelFunc: bJ };
function vJ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s, o = r.shape.length, u = [], l = w.parseAxisParam(a, r.shape), c = l, p = C.getAxesPermutation(c, o), d = r;
  p != null && (d = _t({ inputs: { x: r }, backend: n, attrs: { perm: p } }), c = C.getInnerMostAxes(c.length, o), u.push(d)), C.assertAxesAreInnerMostDims("prod", c, o);
  let h;
  if (n.shouldExecuteOnCPU([d])) {
    let f = n.texData.get(d.dataId).values, { outVals: m, outShape: g, outDtype: b } = IX(d.shape, d.dtype, f, c);
    h = n.makeTensorInfo(g, b, m);
  } else {
    let [f, m] = C.computeOutAndReduceShapes(d.shape, c), g = w.sizeFromShape(m), b = he({ inputs: { x: d }, backend: n, attrs: { shape: [-1, g] } }), y = bp(r.dtype), v = Si(b, y, "prod", n);
    h = he({ inputs: { x: v }, backend: n, attrs: { shape: f } }), u.push(b), u.push(v);
  }
  if (i) {
    u.push(h);
    let f = C.expandShapeToKeepDim(h.shape, l);
    h = he({ inputs: { x: h }, backend: n, attrs: { shape: f } });
  }
  return u.forEach((f) => n.disposeIntermediateTensorInfo(f)), h;
}
var xJ = { kernelName: ni, backendName: "webgl", kernelFunc: vJ };
var k2 = (e) => {
  let { backend: t, attrs: n } = e, { start: s, stop: r, step: a, dtype: i } = n, o = CX(s, r, a, i);
  return t.makeTensorInfo([o.length], i, o);
};
var wJ = { kernelName: Nl, backendName: "webgl", kernelFunc: k2 };
var kJ = "return 1.0 / x;";
var SJ = Ke({ opSnippet: kJ });
var IJ = { kernelName: Tl, backendName: "webgl", kernelFunc: SJ };
var CJ = ss + `
  return (x < 0.0) ? 0.0 : x;
`;
var NJ = `
  vec4 result = x * vec4(greaterThanEqual(x, vec4(0.0)));
  bvec4 isNaN = isnan(x);

  result.r = isNaN.r ? x.r : result.r;
  result.g = isNaN.g ? x.g : result.g;
  result.b = isNaN.b ? x.b : result.b;
  result.a = isNaN.a ? x.a : result.a;

  return result;
`;
var TJ = Ke({ opSnippet: CJ, packedOpSnippet: NJ });
var $J = { kernelName: si, backendName: "webgl", kernelFunc: TJ };
var _J = ss + `
  return (x < 0.0) ? 0.0 : min(6.0, x);
`;
var AJ = `
  vec4 result = min(x, vec4(6.)) * vec4(greaterThanEqual(x, vec4(0.0)));
  bvec4 isNaN = isnan(x);

  result.r = isNaN.r ? x.r : result.r;
  result.g = isNaN.g ? x.g : result.g;
  result.b = isNaN.b ? x.b : result.b;
  result.a = isNaN.a ? x.a : result.a;

  return result;
`;
var EJ = Ke({ opSnippet: _J, packedOpSnippet: AJ });
var RJ = { kernelName: ai, backendName: "webgl", kernelFunc: EJ };
var DJ = class {
  constructor(e, t, n, s, r) {
    this.variableNames = ["A"], this.outputShape = [];
    let [a, i, o, u] = e;
    this.outputShape = [a, t, n, u];
    let l = [s && t > 1 ? i - 1 : i, s && n > 1 ? o - 1 : o], c = [s && t > 1 ? t - 1 : t, s && n > 1 ? n - 1 : n], p;
    r ? p = "(vec2(yRC) + vec2(0.5)) * effectiveInputOverOutputRatioRC - vec2(0.5)" : p = "vec2(yRC) * effectiveInputOverOutputRatioRC", this.userCode = `
      const vec2 effectiveInputOverOutputRatioRC = vec2(
          ${l[0] / c[0]},
          ${l[1] / c[1]});
      const vec2 inputShapeRC = vec2(${i}.0, ${o}.0);

      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int d = coords[3];
        ivec2 yRC = coords.yz;

        // Fractional source index.
        vec2 sourceFracIndexRC = ${p};

        // Compute the four integer indices.
        ivec2 sourceFloorRC = ivec2(max(sourceFracIndexRC, vec2(0.0)));
        ivec2 sourceCeilRC = ivec2(
          min(inputShapeRC - 1.0, ceil(sourceFracIndexRC)));

        float topLeft = getA(b, sourceFloorRC.x, sourceFloorRC.y, d);
        float bottomLeft = getA(b, sourceCeilRC.x, sourceFloorRC.y, d);
        float topRight = getA(b, sourceFloorRC.x, sourceCeilRC.y, d);
        float bottomRight = getA(b, sourceCeilRC.x, sourceCeilRC.y, d);

        vec2 fracRC = sourceFracIndexRC - vec2(sourceFloorRC);

        float top = topLeft + (topRight - topLeft) * fracRC.y;
        float bottom = bottomLeft + (bottomRight - bottomLeft) * fracRC.y;
        float newValue = top + (bottom - top) * fracRC.x;

        setOutput(newValue);
      }
    `;
  }
};
var FJ = class {
  constructor(e, t, n, s, r) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true, this.outputShape = [];
    let [a, i, o, u] = e;
    this.outputShape = [a, t, n, u];
    let l = [s && t > 1 ? i - 1 : i, s && n > 1 ? o - 1 : o], c = [s && t > 1 ? t - 1 : t, s && n > 1 ? n - 1 : n], p;
    r ? p = "(vec3(yRC) + vec3(0.5)) * effectiveInputOverOutputRatioRC - vec3(0.5)" : p = "vec3(yRC) * effectiveInputOverOutputRatioRC", this.userCode = `
      const vec3 effectiveInputOverOutputRatioRC = vec3(
          ${l[0] / c[0]},
          ${l[1] / c[1]},
          ${l[1] / c[1]});
      const vec3 inputShapeRC = vec3(${i}.0, ${o}.0,
                                     ${o}.0);

      float getAValue(int b, int r, int c, int d) {
        return getChannel(getA(b, r, c, d), vec2(c, d));
      }

      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int d = coords[3];
        // Calculate values for next column in yRC.z.
        ivec3 yRC = coords.yzz + ivec3(0, 0, 1);

        // Fractional source index.
        vec3 sourceFracIndexRC = ${p};

        // Compute the four integer indices.
        ivec3 sourceFloorRC = ivec3(max(sourceFracIndexRC, vec3(0.0)));
        ivec3 sourceCeilRC = ivec3(
          min(inputShapeRC - 1.0, ceil(sourceFracIndexRC)));

        // Should we calculate next column and row elements in 2x2 packed cell.
        bool hasNextCol = d < ${u - 1};
        bool hasNextRow = coords.z < ${n - 1};

        // In parallel, construct four corners for all four components in
        // packed 2x2 cell.
        vec4 topLeft = vec4(
          getAValue(b, sourceFloorRC.x, sourceFloorRC.y, d),
          hasNextCol ? getAValue(b, sourceFloorRC.x, sourceFloorRC.y, d + 1)
                     : 0.0,
          hasNextRow ? getAValue(b, sourceFloorRC.x, sourceFloorRC.z, d)
                     : 0.0,
          (hasNextRow && hasNextCol) ?
            getAValue(b, sourceFloorRC.x, sourceFloorRC.z, d + 1) : 0.0);

        vec4 bottomLeft = vec4(
          getAValue(b, sourceCeilRC.x, sourceFloorRC.y, d),
          hasNextCol ? getAValue(b, sourceCeilRC.x, sourceFloorRC.y, d + 1)
                     : 0.0,
          hasNextRow ? getAValue(b, sourceCeilRC.x, sourceFloorRC.z, d)
                     : 0.0,
          (hasNextRow && hasNextCol) ?
            getAValue(b, sourceCeilRC.x, sourceFloorRC.z, d + 1) : 0.0);

        vec4 topRight = vec4(
          getAValue(b, sourceFloorRC.x, sourceCeilRC.y, d),
          hasNextCol ? getAValue(b, sourceFloorRC.x, sourceCeilRC.y, d + 1)
                     : 0.0,
          hasNextRow ? getAValue(b, sourceFloorRC.x, sourceCeilRC.z, d)
                     : 0.0,
          (hasNextRow && hasNextCol) ?
            getAValue(b, sourceFloorRC.x, sourceCeilRC.z, d + 1) : 0.0);

        vec4 bottomRight = vec4(
          getAValue(b, sourceCeilRC.x, sourceCeilRC.y, d),
          hasNextCol ? getAValue(b, sourceCeilRC.x, sourceCeilRC.y, d + 1)
                     : 0.0,
          hasNextRow ? getAValue(b, sourceCeilRC.x, sourceCeilRC.z, d)
                     : 0.0,
          (hasNextRow && hasNextCol) ?
            getAValue(b, sourceCeilRC.x, sourceCeilRC.z, d + 1) : 0.0);

        vec3 fracRC = sourceFracIndexRC - vec3(sourceFloorRC);

        vec4 top = mix(topLeft, topRight, fracRC.yyzz);
        vec4 bottom = mix(bottomLeft, bottomRight, fracRC.yyzz);
        vec4 newValue = mix(top, bottom, fracRC.x);

        setOutput(newValue);
      }
    `;
  }
};
function OJ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r } = t, { alignCorners: a, halfPixelCenters: i, size: o } = s, [u, l] = o, c = K().getBool("WEBGL_PACK_IMAGE_OPERATIONS") ? new FJ(r.shape, u, l, a, i) : new DJ(r.shape, u, l, a, i);
  return n.runWebGLProgram(c, [r], "float32");
}
var PJ = { kernelName: ri, backendName: "webgl", kernelFunc: OJ };
var zJ = class {
  constructor(e, t, n) {
    this.variableNames = ["dy"], this.outputShape = [], this.outputShape = t;
    let [, s, r] = t, [, a, i] = e, o = [n && a > 1 ? s - 1 : s, n && i > 1 ? r - 1 : r], u = [n && a > 1 ? a - 1 : a, n && i > 1 ? i - 1 : i], l = o[0] / u[0], c = o[1] / u[1], p = 1 / l, d = 1 / c, h = Math.ceil(p) * 2 + 2, f = Math.ceil(d) * 2 + 2;
    this.userCode = `
      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int d = coords[3];
        int r = coords[1];
        int c = coords[2];

        float accumulator = 0.0;

        const float heightScale = float(${l});
        const float widthScale = float(${c});

        const float invHeightScale = float(${p});
        const float invWidthScale = float(${d});

        const int winHeight = int(${h});
        const int winWidth = int(${f});

        // Compute bounds for where in dy we will look
        float startRLerp = floor(float(r) * invHeightScale);
        int startDyR = int(startRLerp - float(winHeight / 2));

        float startCLerp = floor(float(c) * invWidthScale);
        int startDyC = int(startCLerp - float(winWidth / 2));

        // Loop over dy
        for (int dyROffset = 0; dyROffset < winHeight; dyROffset++) {
          int dyR = dyROffset + startDyR;

          // Guard against the window exceeding the bounds of dy
          if (dyR < 0 || dyR >= ${a}) {
            continue;
          }

          for (int dyCOffset = 0; dyCOffset < winWidth; dyCOffset++) {
            int dyC = dyCOffset + startDyC;

            // Guard against the window exceeding the bounds of dy
            if (dyC < 0 || dyC >= ${i}) {
              continue;
            }

            float dxR = float(dyR) * heightScale;
            int topDxRIndex = int(floor(dxR));
            int bottomDxRIndex = int(min(ceil(dxR), ${s - 1}.0));
            float dxRLerp = dxR - float(topDxRIndex);
            float inverseDxRLerp = 1.0 - dxRLerp;

            float dxC = float(dyC) * widthScale;
            int leftDxCIndex = int(floor(dxC));
            int rightDxCIndex = int(min(ceil(dxC), ${r - 1}.0));
            float dxCLerp = dxC - float(leftDxCIndex);
            float inverseDxCLerp = 1.0 - dxCLerp;

            if (r == topDxRIndex && c == leftDxCIndex) {
              // topLeft
              accumulator +=
                getDy(b, dyR, dyC, d) * inverseDxRLerp * inverseDxCLerp;
            }

            if (r == topDxRIndex && c == rightDxCIndex) {
              // topRight
              accumulator += getDy(b, dyR, dyC, d) * inverseDxRLerp * dxCLerp;
            }

            if (r == bottomDxRIndex && c == leftDxCIndex) {
              // bottomLeft
              accumulator += getDy(b, dyR, dyC, d) * dxRLerp * inverseDxCLerp;
            }

            if (r == bottomDxRIndex && c == rightDxCIndex) {
              // bottomRight
              accumulator += getDy(b, dyR, dyC, d) * dxRLerp * dxCLerp;
            }
          }
        }
        // End loop over dy

        setOutput(accumulator);
      }
    `;
  }
};
function MJ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r, dy: a } = t, { alignCorners: i } = s, o = new zJ(a.shape, r.shape, i);
  return n.runWebGLProgram(o, [a], a.dtype);
}
var LJ = { kernelName: Dg, backendName: "webgl", kernelFunc: MJ };
var BJ = class {
  constructor(e, t, n, s, r) {
    this.variableNames = ["A"], this.outputShape = [];
    let [a, i, o, u] = e;
    this.outputShape = [a, t, n, u];
    let l = [s && t > 1 ? i - 1 : i, s && n > 1 ? o - 1 : o], c = [s && t > 1 ? t - 1 : t, s && n > 1 ? n - 1 : n], p = s ? "0.5" : "0.0", d;
    r ? d = "max((vec2(yRC) + vec2(0.5)) * effectiveInputOverOutputRatioRC, vec2(0.0))" : d = "vec2(yRC) * effectiveInputOverOutputRatioRC", this.userCode = `
      const vec2 effectiveInputOverOutputRatioRC = vec2(
          ${l[0] / c[0]},
          ${l[1] / c[1]});
      const vec2 inputShapeRC = vec2(${i}.0, ${o}.0);

      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int d = coords[3];
        ivec2 yRC = coords.yz;

        // Fractional source index.
        vec2 sourceFracIndexRC = ${d};

        // Compute the coordinators of nearest neighbor point.
        ivec2 sourceNearestRC = ivec2(
          min(inputShapeRC - 1.0, floor(sourceFracIndexRC + ${p})));
        float newValue = getA(b, sourceNearestRC.x, sourceNearestRC.y, d);

        setOutput(newValue);
      }
    `;
  }
};
var VJ = class {
  constructor(e, t, n, s, r) {
    this.variableNames = ["A"], this.packedInputs = true, this.packedOutput = true, this.outputShape = [];
    let [a, i, o, u] = e;
    this.outputShape = [a, t, n, u];
    let l = [s && t > 1 ? i - 1 : i, s && n > 1 ? o - 1 : o], c = [s && t > 1 ? t - 1 : t, s && n > 1 ? n - 1 : n], p = s ? "0.5" : "0.0", d;
    r ? d = "max((vec3(yRC) + vec3(0.5)) * effectiveInputOverOutputRatioRC, vec3(0.0))" : d = "vec3(yRC) * effectiveInputOverOutputRatioRC", this.userCode = `
      const vec3 effectiveInputOverOutputRatioRC = vec3(
          ${l[0] / c[0]},
          ${l[1] / c[1]},
          ${l[1] / c[1]});
      const vec3 inputShapeRC = vec3(${i}.0, ${o}.0,
                                     ${o}.0);

      float getAValue(int b, int r, int c, int d) {
        return getChannel(getA(b, r, c, d), vec2(c, d));
      }

      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int d = coords[3];
        // Calculate values for next column in yRC.z.
        ivec3 yRC = coords.yzz + ivec3(0, 0, 1);

        // Fractional source index.
        vec3 sourceFracIndexRC = ${d};

        // Compute the coordinators of nearest neighbor point.
        ivec3 sourceNearestRC = ivec3(
          min(inputShapeRC - 1.0, floor(sourceFracIndexRC + ${p})));

        // Should we calculate next column and row elements in 2x2 packed cell.
        bool hasNextCol = d < ${u - 1};
        bool hasNextRow = coords.z < ${n - 1};

        vec4 newValue = vec4(
          getAValue(b, sourceNearestRC.x, sourceNearestRC.y, d),
          hasNextCol ? getAValue(b, sourceNearestRC.x, sourceNearestRC.y, d + 1)
                     : 0.0,
          hasNextRow ? getAValue(b, sourceNearestRC.x, sourceNearestRC.z, d)
                     : 0.0,
          (hasNextRow && hasNextCol) ?
            getAValue(b, sourceNearestRC.x, sourceNearestRC.z, d + 1) : 0.0);

        setOutput(newValue);
      }
    `;
  }
};
function WJ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r } = t, { alignCorners: a, halfPixelCenters: i, size: o } = s, [u, l] = o, c = K().getBool("WEBGL_PACK_IMAGE_OPERATIONS") ? new VJ(r.shape, u, l, a, i) : new BJ(r.shape, u, l, a, i);
  return n.runWebGLProgram(c, [r], r.dtype);
}
var UJ = { kernelName: $l, backendName: "webgl", kernelFunc: WJ };
var GJ = class {
  constructor(e, t, n) {
    this.variableNames = ["dy"], this.outputShape = [], this.outputShape = t;
    let [, s, r] = t, [, a, i] = e, o = [n && a > 1 ? s - 1 : s, n && i > 1 ? r - 1 : r], u = [n && a > 1 ? a - 1 : a, n && i > 1 ? i - 1 : i], l = o[0] / u[0], c = o[1] / u[1], p = 1 / l, d = 1 / c, h = Math.ceil(p) * 2 + 2, f = Math.ceil(d) * 2 + 2;
    this.userCode = `
      void main() {
        ivec4 coords = getOutputCoords();
        int b = coords[0];
        int d = coords[3];
        int r = coords[1];
        int c = coords[2];

        float accumulator = 0.0;

        const float heightScale = float(${l});
        const float widthScale = float(${c});

        const float invHeightScale = float(${p});
        const float invWidthScale = float(${d});

        const int winHeight = int(${h});
        const int winWidth = int(${f});

        // Compute bounds for where in dy we will look
        float startRLerp = floor(float(r) * invHeightScale);
        int startDyR = int(floor(startRLerp - float(winHeight / 2)));

        float startCLerp = floor(float(c) * invWidthScale);
        int startDyC = int(floor(startCLerp - float(winWidth / 2)));

        // Loop over dy
        for (int dyROffset = 0; dyROffset < winHeight; dyROffset++) {
          int dyR = dyROffset + startDyR;

          // Guard against the window exceeding the bounds of dy
          if (dyR < 0 || dyR >= ${a}) {
            continue;
          }

          for (int dyCOffset = 0; dyCOffset < winWidth; dyCOffset++) {
            int dyC = dyCOffset + startDyC;

            // Guard against the window exceeding the bounds of dy
            if (dyC < 0 || dyC >= ${i}) {
              continue;
            }

            float sourceFracRow =
              float(${o[0]}) *
                (float(dyR) / float(${u[0]}));

            float sourceFracCol =
                float(${o[1]}) *
                  (float(dyC) / float(${u[1]}));

            int sourceNearestRow = int(min(
                float(int(${s}) - 1),
                ${n} ? float(round(sourceFracRow)) :
                                  float(floor(sourceFracRow))));

            int sourceNearestCol = int(min(
                float(int(${r}) - 1),
                ${n} ? float(round(sourceFracCol)) :
                                  float(floor(sourceFracCol))));

            if (r == sourceNearestRow && c == sourceNearestCol) {
              accumulator += getDy(b, dyR, dyC, d);
            }
          }
        }
        // End loop over dy

        setOutput(accumulator);
      }
    `;
  }
};
function HJ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r, dy: a } = t, { alignCorners: i } = s, o = new GJ(a.shape, r.shape, i);
  return n.runWebGLProgram(o, [a], a.dtype);
}
var qJ = { kernelName: Rg, backendName: "webgl", kernelFunc: HJ };
var jJ = class {
  constructor(e, t) {
    this.variableNames = ["x"];
    let n = e.length;
    if (n > 4)
      throw new Error(`WebGL backend: Reverse of rank-${n} tensor is not yet supported`);
    if (this.outputShape = e, n === 1) {
      this.userCode = `
        void main() {
          int coord = getOutputCoords();
          setOutput(getX(${e[0]} - coord - 1));
        }
      `;
      return;
    }
    let s = (i) => t.indexOf(i) !== -1 && e[i] !== 1 ? `${e[i]} - coords[${i}] - 1` : `coords[${i}]`, r = e.map((i, o) => s(o)).join(","), a = ot(n);
    this.userCode = `
      void main() {
        ${a} coords = getOutputCoords();
        setOutput(getX(${r}));
      }
    `;
  }
};
var KJ = class {
  constructor(e, t) {
    this.variableNames = ["x"], this.packedInputs = true, this.packedOutput = true;
    let n = e.length;
    if (n > 4)
      throw new Error(`WebGL backend: Reverse of rank-${n} tensor is not yet supported`);
    this.outputShape = e;
    let s = ln("rc", n), r = `${s[n - 1]} + 1 < ${this.outputShape[n - 1]}`, a = `${s[n - 2]} + 1 < ${this.outputShape[n - 2]}`, i = ot(n);
    n === 1 ? this.userCode = `
        void main(){
          int rc = getOutputCoords();
          vec4 result = vec4(0.);
          result.r = getChannel(getX(${e[0]} - rc - 1),
            ${e[0]} - rc - 1);
          if(${r}){
              result.g = getChannel(getX(${e[0]} - (rc  + 1) - 1),
                ${e[0]} - (rc  + 1) - 1);
          }
          setOutput(result);
        }
      ` : this.userCode = `
        void main() {
          ${i} rc = getOutputCoords();
          vec4 result = vec4(0.);
          result.r = ${o(s.slice())};
          if(${r}){
            result.g = ${u(s.slice())};
          }
          if(${a}) {
            result.b = ${l(s.slice())};
            if(${r}) {
              result.a = ${c(s.slice())};
            }
          }
          setOutput(result);
        }
    `;
    function o(h) {
      return p(h);
    }
    function u(h) {
      return h[n - 1] = "(" + h[n - 1] + " + 1)", p(h);
    }
    function l(h) {
      return h[n - 2] = "(" + h[n - 2] + " + 1)", p(h);
    }
    function c(h) {
      return h[n - 1] = "(" + h[n - 1] + " + 1)", h[n - 2] = "(" + h[n - 2] + " + 1)", p(h);
    }
    function p(h) {
      let f = e.map((b, y) => d(y, h)), m = f.join(","), g = f.slice(-2).join(",");
      return `getChannel(getX(${m}), vec2(${g}))`;
    }
    function d(h, f) {
      return t.indexOf(h) !== -1 && e[h] !== 1 ? `${e[h]} - ${f[h]} - 1` : `${f[h]}`;
    }
  }
};
function XJ(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { dims: a } = s, i = r.shape.length, o = w.parseAxisParam(a, r.shape);
  if (i === 0)
    return Rn({ inputs: { x: r }, backend: n });
  let u = K().getBool("WEBGL_PACK_ARRAY_OPERATIONS") ? new KJ(r.shape, o) : new jJ(r.shape, o);
  return n.runWebGLProgram(u, [r], r.dtype);
}
var YJ = { kernelName: Po, backendName: "webgl", kernelFunc: XJ };
var QJ = class {
  constructor(e, t) {
    this.variableNames = ["Image"], this.outputShape = [], this.customUniforms = [{ name: "params", type: "vec4" }];
    let n = e[1], s = e[2];
    this.outputShape = e;
    let r = "";
    typeof t == "number" ? r = `float outputValue = ${t.toFixed(2)};` : r = `
        vec3 fill = vec3(${t.join(",")});
        float outputValue = fill[coords[3]];`, this.userCode = `
        void main() {
          ivec4 coords = getOutputCoords();
          int x = coords[2];
          int y = coords[1];
          float coordXFloat = (float(x) - params[0]) * params[3] -
            (float(y) - params[1]) * params[2];
          float coordYFloat = (float(x) - params[0]) * params[2] +
            (float(y) - params[1]) * params[3];
          int coordX = int(round(coordXFloat + params[0]));
          int coordY = int(round(coordYFloat + params[1]));
          ${r}
          if(coordX >= 0 && coordX < ${s} && coordY >= 0 && coordY < ${n}) {
            outputValue = getImage(coords[0], coordY, coordX, coords[3]);
          }
          setOutput(outputValue);
        }
    `;
  }
};
var ZJ = { kernelName: Yo, backendName: "webgl", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { image: s } = e, { radians: r, fillValue: a, center: i } = t, o = n, u = new QJ(s.shape, a), [l, c] = C.getImageCenter(i, s.shape[1], s.shape[2]), p = [[l, c, Math.sin(r), Math.cos(r)]];
  return o.runWebGLProgram(u, [s], s.dtype, p);
} };
var JJ = `
  // OpenGL ES does not support round function.
  // The algorithm is based on banker's rounding.
  float base = floor(x);
  if ((x - base) < 0.5) {
    return floor(x);
  } else if ((x - base) > 0.5) {
    return ceil(x);
  } else {
    if (mod(base, 2.0) == 0.0) {
      return base;
    } else {
      return base + 1.0;
    }
  }
`;
var eee = Ke({ opSnippet: JJ });
var tee = { kernelName: zo, backendName: "webgl", kernelFunc: eee };
var nee = "return inversesqrt(x);";
var see = Ke({ opSnippet: nee, cpuKernelImpl: NX });
var ree = { kernelName: ii, backendName: "webgl", kernelFunc: see };
var S2 = class {
  constructor(e, t, n, s, r, a, i = true) {
    this.variableNames = ["updates", "indices", "defaultValue"], this.outputShape = a;
    let o = ot(r.length), u = ot(a.length), l = "";
    n === 1 ? l = "i" : n === 2 && (l = "i, j");
    let c = `getIndices(${l})`, p = "";
    s === 1 ? p = "i" : s === 2 && (p = "i, coords[1]");
    let d = `getUpdates(${p})`, h = t > 1 ? "strides[j]" : "strides";
    this.userCode = `
        ${o} strides = ${o}(${r});

        void main() {
          ${u} coords = getOutputCoords();
          float sum = 0.0;
          bool found = false;
          for (int i = 0; i < ${e}; i++) {
            int flattenedIndex = 0;
            for (int j = 0; j < ${t}; j++) {
              int index = round(${c});
              flattenedIndex += index * ${h};
            }
            if (flattenedIndex == coords[0]) {
              sum += ${d};
              found = true;
            }
          }
          setOutput(mix(getDefaultValue(), sum, float(found)));
        }
      `;
  }
};
function aee(e) {
  let { inputs: t, backend: n, attrs: s } = e, { indices: r, updates: a } = t, { shape: i } = s, { sliceRank: o, numUpdates: u, sliceSize: l, strides: c, outputSize: p } = C.calculateShapes(a, r, i), d = [p / l, l];
  if (p === 0)
    return n.makeTensorInfo(i, r.dtype);
  let h = he({ inputs: { x: r }, backend: n, attrs: { shape: [u, o] } }), f = he({ inputs: { x: a }, backend: n, attrs: { shape: [u, l] } }), m = n.makeTensorInfo([], "float32", new Float32Array([0])), g = new S2(u, o, h.shape.length, f.shape.length, c, d), b = n.runWebGLProgram(g, [f, h, m], f.dtype), y = he({ inputs: { x: b }, backend: n, attrs: { shape: i } });
  return n.disposeIntermediateTensorInfo(h), n.disposeIntermediateTensorInfo(f), n.disposeIntermediateTensorInfo(b), n.disposeIntermediateTensorInfo(m), y;
}
var iee = { kernelName: Mo, backendName: "webgl", kernelFunc: aee };
var oee = class {
  constructor(e, t, n, s) {
    this.variableNames = ["sortedSequence", "values"], this.customUniforms = [{ name: "numInputs", type: "int" }], this.outputShape = [e, n];
    let r = "while (left < right) {", a = `for (int i = 0; i < ${Math.ceil(Math.log2(t + 1))}; ++i) { if (left >= right) break;`, i = K().getNumber("WEBGL_VERSION") === 2 ? r : a, o = s === "left" ? "<" : "<=";
    this.userCode = `
       int findBound(int batch, float value) {
         int left = 0;
         int right = numInputs;
         int mid;
         ${i}
           mid = (left + right) / 2;
           if (getSortedSequence(batch, mid) ${o} value) {
             left = mid + 1;
           } else {
             right = mid;
           }
         }
         return right;
       }

       void main() {
         ivec2 coords = getOutputCoords();
         int batch = coords[0];
         int valueIndex = coords[1];

         float value = getValues(batch, valueIndex);

         setOutput(float(findBound(batch, value)));
       }
     `;
  }
};
function uee(e) {
  let { inputs: t, backend: n, attrs: s } = e, { sortedSequence: r, values: a } = t, { side: i } = s, o = new oee(r.shape[0], r.shape[1], a.shape[1], i), u = [[r.shape[1]]];
  return n.runWebGLProgram(o, [r, a], "int32", u);
}
var lee = { kernelName: Fg, backendName: "webgl", kernelFunc: uee };
var cee = class {
  constructor(e, t, n) {
    this.variableNames = ["c", "a", "b"], this.outputShape = t;
    let s, r;
    if (n > 4)
      throw Error(`Where for rank ${n} is not yet supported`);
    if (n === 1)
      r = "resRC", s = "resRC";
    else {
      let i = ["resRC.x", "resRC.y", "resRC.z", "resRC.w"], o = [], u = [];
      for (let l = 0; l < t.length; l++)
        u.push(`${i[l]}`), l < e && o.push(`${i[l]}`);
      s = o.join(), r = u.join();
    }
    let a = ot(n);
    this.userCode = `
      void main() {
        ${a} resRC = getOutputCoords();
        float cVal = getC(${s});
        if (cVal >= 1.0) {
          setOutput(getA(${r}));
        } else {
          setOutput(getB(${r}));
        }
      }
    `;
  }
};
function dee(e) {
  let { inputs: t, backend: n } = e, { condition: s, t: r, e: a } = t, i = new cee(s.shape.length, r.shape, r.shape.length);
  return n.runWebGLProgram(i, [s, r, a], cn(r.dtype, a.dtype));
}
var pee = { kernelName: Lo, backendName: "webgl", kernelFunc: dee };
var hee = `
  // Stable and Attracting Fixed Point (0, 1) for Normalized Weights.
  // see: https://arxiv.org/abs/1706.02515
  float scaleAlpha = ${C.SELU_SCALEALPHA};
  float scale = ${C.SELU_SCALE};
  return (x >= 0.0) ? scale * x : scaleAlpha * (exp(x) - 1.0);
`;
var fee = Ke({ opSnippet: hee });
var mee = { kernelName: _l, backendName: "webgl", kernelFunc: fee };
var gee = du + `
  return 1.0 / (1.0 + exp(-1.0 * x));
`;
var bee = `
  vec4 result = 1.0 / (1.0 + exp(-1.0 * x));
  bvec4 isNaN = isnan(x);

  result.r = isNaN.r ? x.r : result.r;
  result.g = isNaN.g ? x.g : result.g;
  result.b = isNaN.b ? x.b : result.b;
  result.a = isNaN.a ? x.a : result.a;

  return result;
`;
var yee = Ke({ opSnippet: gee, packedOpSnippet: bee, cpuKernelImpl: $X });
var vee = { kernelName: ui, backendName: "webgl", kernelFunc: yee };
var xee = `
  if (isnan(x)) { return 0.0; }
  return sign(x);
`;
var wee = Ke({ opSnippet: xee });
var kee = { kernelName: Al, backendName: "webgl", kernelFunc: wee };
var See = du + `
  return sin(x);
`;
var Iee = Ke({ opSnippet: See });
var Cee = { kernelName: oi, backendName: "webgl", kernelFunc: Iee };
var Nee = `
  float e2x = exp(x);
  return (e2x - 1.0 / e2x) / 2.0;
`;
var Tee = Ke({ opSnippet: Nee });
var $ee = { kernelName: Vo, backendName: "webgl", kernelFunc: Tee };
var _ee = `
  float epsilon = 1.1920928955078125e-7;
  float threshold = log(epsilon) + 2.0;

  bool too_large = x > -threshold;
  bool too_small = x < threshold;

  float result;
  float exp_x = exp(x);

  if (too_large){
    result = x;
  }
  else if (too_small){
    result = exp_x;
  }
  else{
    result = log(exp_x + 1.0);
  }
  return result;
`;
var Aee = Ke({ opSnippet: _ee });
var Eee = { kernelName: El, backendName: "webgl", kernelFunc: Aee };
var Ree = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockShape: a, paddings: i } = s;
  w.assert(r.shape.length <= 4, () => "spaceToBatchND for rank > 4 with a WebGL backend not implemented yet");
  let o = a.reduce((b, y) => b * y), u = [[0, 0]];
  u.push(...i);
  for (let b = 1 + a.length; b < r.shape.length; ++b)
    u.push([0, 0]);
  let l = [], c = w2({ inputs: { x: r }, backend: n, attrs: { paddings: u, constantValue: 0 } }), p = C.getReshaped(c.shape, a, o, false), d = C.getPermuted(p.length, a.length, false), h = C.getReshapedPermuted(c.shape, a, o, false), f = he({ inputs: { x: c }, backend: n, attrs: { shape: p } }), m = _t({ inputs: { x: f }, backend: n, attrs: { perm: d } }), g = he({ inputs: { x: m }, backend: n, attrs: { shape: h } });
  return l.push(c), l.push(f), l.push(m), l.forEach((b) => n.disposeIntermediateTensorInfo(b)), g;
};
var Dee = { kernelName: Wo, backendName: "webgl", kernelFunc: Ree };
function Fee(e) {
  let { inputs: t, backend: n } = e, { indices: s, values: r, denseShape: a, defaultValue: i } = t;
  if (a.shape.length !== 1)
    throw new Error(`Dense shape must be a vector, saw:
         ${a.shape}`);
  if (s.shape.length !== 2)
    throw new Error(`Indices must be a matrix, saw:
         ${s.shape}`);
  if (r.shape.length !== 1)
    throw new Error(`Values must be a vector, saw:
         ${r.shape}`);
  if (i.shape.length !== 0)
    throw new Error(`Default value must be a scalar, saw:
        ${i.shape}`);
  let o = n.readSync(s.dataId), u = n.readSync(r.dataId), l = n.readSync(a.dataId), c = n.readSync(i.dataId)[0], [p, d, h, f, m] = AX(o, s.shape, s.dtype, u, r.dtype, l, c);
  return [n.makeTensorInfo(d, s.dtype, p), n.makeTensorInfo([d[0]], r.dtype, h), n.makeTensorInfo([f.length], "bool", new Uint8Array(f.map((g) => Number(g)))), n.makeTensorInfo([m.length], s.dtype, new Int32Array(m))];
}
var Oee = { kernelName: cp, backendName: "webgl", kernelFunc: Fee };
function Pee(e) {
  let { inputs: t, backend: n } = e, { inputIndices: s, inputShape: r, newShape: a } = t;
  if (s.shape.length !== 2)
    throw new Error(`Input indices should be a matrix but received shape ${s.shape}`);
  if (r.shape.length !== 1)
    throw new Error(`Input shape should be a vector but received shape ${r.shape}`);
  if (a.shape.length !== 1)
    throw new Error(`Target shape should be a vector but received shape ${a.shape}`);
  let i = Array.from(n.readSync(r.dataId)), o = n.readSync(s.dataId), u = Array.from(n.readSync(a.dataId)), [l, c, p] = EX(o, s.shape, s.dtype, i, u);
  return [n.makeTensorInfo(c, s.dtype, l), n.makeTensorInfo([p.length], a.dtype, new Int32Array(p))];
}
var zee = { kernelName: Rl, backendName: "webgl", kernelFunc: Pee };
function Mee(e) {
  let { inputs: t, backend: n } = e, { data: s, indices: r, segmentIds: a } = t;
  if (s.shape.length < 1)
    throw new Error("Data should be at least 1 dimensional but received scalar");
  if (r.shape.length !== 1)
    throw new Error(`Indices should be a vector but received shape
              ${r.shape}`);
  if (a.shape.length !== 1)
    throw new Error(`Segment ids should be a vector but received shape
              ${a.shape}`);
  let i = n.readSync(s.dataId), o = n.readSync(r.dataId), u = n.readSync(a.dataId), [l, c] = G1(i, s.shape, s.dtype, o, u, true);
  return n.makeTensorInfo(c, s.dtype, l);
}
var Lee = { kernelName: dp, backendName: "webgl", kernelFunc: Mee };
function Bee(e) {
  let { inputs: t, backend: n } = e, { data: s, indices: r, segmentIds: a } = t;
  if (s.shape.length < 1)
    throw new Error("Data should be at least 1 dimensional but received scalar");
  if (r.shape.length !== 1)
    throw new Error(`Indices should be a vector but received shape
             ${r.shape}`);
  if (a.shape.length !== 1)
    throw new Error(`Segment ids should be a vector but received shape
             ${a.shape}`);
  let i = n.readSync(s.dataId), o = n.readSync(r.dataId), u = n.readSync(a.dataId), [l, c] = G1(i, s.shape, s.dtype, o, u);
  return n.makeTensorInfo(c, s.dtype, l);
}
var Vee = { kernelName: pp, backendName: "webgl", kernelFunc: Bee };
function Wee(e) {
  let { inputs: t, backend: n, attrs: s } = e, { sparseIndices: r, sparseValues: a, defaultValue: i } = t, { outputShape: o } = s, { sliceRank: u, numUpdates: l, sliceSize: c, strides: p, outputSize: d } = C.calculateShapes(a, r, o), h = false;
  if (a.dtype === "string") {
    let b = n.bufferSync(r), y = n.bufferSync(a), v = w.decodeString(n.readSync(i.dataId)[0]), x = TX(b, y, o, d, c, l, u, p, v, h);
    return n.makeTensorInfo(o, x.dtype, x.values);
  }
  let f = new S2(l, u, r.shape.length, a.shape.length, p, [d, 1], h), m = n.runWebGLProgram(f, [a, r, i], a.dtype), g = he({ inputs: { x: m }, backend: n, attrs: { shape: o } });
  return n.disposeIntermediateTensorInfo(m), g;
}
var Uee = { kernelName: hp, backendName: "webgl", kernelFunc: Wee };
function Gee(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { numOrSizeSplits: a, axis: i } = s, o = w.parseAxisParam(i, r.shape)[0], u = C.prepareSplitSize(r, a, o), l = r.shape.length, c = new Array(l).fill(0), p = r.shape.slice();
  return u.map((d) => {
    let h = [...p];
    h[o] = d;
    let f = pu({ inputs: { x: r }, backend: n, attrs: { begin: c, size: h } });
    return c[o] += d, f;
  });
}
var Hee = { kernelName: Uo, backendName: "webgl", kernelFunc: Gee };
var Aw = "return sqrt(x);";
var qee = Ke({ opSnippet: Aw, packedOpSnippet: Aw, cpuKernelImpl: RX });
var jee = { kernelName: li, backendName: "webgl", kernelFunc: qee };
var Kee = "return x * x;";
var Xee = Ke({ opSnippet: Kee });
var Yee = { kernelName: Dl, backendName: "webgl", kernelFunc: Xee };
var Ew = "return (a - b) * (a - b);";
var Qee = jt({ opSnippet: Ew, packedOpSnippet: Ew });
var Zee = { kernelName: pi, backendName: "webgl", kernelFunc: Qee };
function Jee({ inputs: e, attrs: t, backend: n }) {
  let { x: s } = e, r = ss + `
    return x > 0.0 ? 1.0 : float(${t.alpha});
  `, a = new Gs(s.shape, r);
  return n.runWebGLProgram(a, [s], s.dtype);
}
var ete = { kernelName: gi, backendName: "webgl", kernelFunc: Jee };
var tte = class {
  constructor(e, t, n) {
    this.variableNames = ["x"], this.outputShape = n;
    let s = n.length, r = ot(n.length), a = ot(n.length), i = "";
    if (s === 1)
      i = "coords * strides + begin";
    else {
      let o = 0;
      i = n.map((u, l) => (o++, n.length === 1 ? `coords * strides[${l}] + begin[${l}]` : `coords[${o - 1}] * strides[${l}] + begin[${l}]`)).join(",");
    }
    this.userCode = `
      ${r} begin = ${r}(${e});
      ${r} strides = ${r}(${t});

      void main() {
        ${a} coords = getOutputCoords();
        setOutput(getX(${i}));
      }
    `;
  }
};
function nte(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { begin: a, end: i, strides: o, beginMask: u, endMask: l, ellipsisMask: c, newAxisMask: p, shrinkAxisMask: d } = s, { finalShapeSparse: h, finalShape: f, isIdentity: m, sliceDim0: g, isSimpleSlice: b, begin: y, end: v, strides: x } = wt.sliceInfo(r.shape, a, i, o, u, l, c, p, d), k;
  if (m)
    k = he({ inputs: { x: r }, backend: n, attrs: { shape: f } });
  else if (g || b) {
    w.assert(r.shape.length >= 1, () => `Input must have rank at least 1, got: ${r.shape.length}`);
    let $ = wt.computeOutShape(y, v, x), R = pu({ inputs: { x: r }, backend: n, attrs: { begin: y, size: $ } });
    k = he({ inputs: { x: R }, backend: n, attrs: { shape: f } }), n.disposeIntermediateTensorInfo(R);
  } else if (n.shouldExecuteOnCPU([r])) {
    let R = n.readSync(r.dataId), E = Ae(r.shape, r.dtype, R), P = DX(h, E, x, y);
    k = n.makeTensorInfo(f, r.dtype, P.values);
  } else {
    let R = new tte(y, x, h);
    k = n.runWebGLProgram(R, [r], r.dtype);
  }
  let I = he({ inputs: { x: k }, backend: n, attrs: { shape: f } });
  return n.disposeIntermediateTensorInfo(k), I;
}
var ste = { kernelName: Go, backendName: "webgl", kernelFunc: nte };
function rte(e) {
  let { inputs: t, backend: n, attrs: s } = e, { separator: r, nGramWidths: a, leftPad: i, rightPad: o, padWidth: u, preserveShortSequences: l } = s, { data: c, dataSplits: p } = t, d = n.readSync(c.dataId), h = n.readSync(p.dataId), [f, m] = FX(d, h, r, a, i, o, u, l);
  return [n.makeTensorInfo([f.length], "string", f), n.makeTensorInfo(p.shape, "int32", m)];
}
var ate = { kernelName: fp, backendName: "webgl", kernelFunc: rte };
function ite(e) {
  let { inputs: t, backend: n, attrs: s } = e, { skipEmpty: r } = s, { input: a, delimiter: i } = t;
  if (a.dtype !== "string")
    throw new Error("Input must be of datatype string");
  if (a.shape.length !== 1)
    throw new Error(`Input must be a vector, got shape: ${a.shape}`);
  if (i.shape.length !== 0)
    throw new Error(`Delimiter must be a scalar, got shape: ${i.shape}`);
  let o = n.readSync(a.dataId), u = n.readSync(i.dataId)[0], [l, c, p] = OX(o, u, r), d = c.length;
  return [n.makeTensorInfo([d, 2], "int32", l), n.makeTensorInfo([d], "string", c), n.makeTensorInfo([2], "int32", new Int32Array(p))];
}
var ote = { kernelName: Og, backendName: "webgl", kernelFunc: ite };
function ute(e) {
  let { inputs: t, backend: n, attrs: s } = e, { numBuckets: r } = s, { input: a } = t;
  if (a.dtype !== "string")
    throw new Error("Input must be of datatype string");
  if (r <= 0)
    throw new Error("Number of buckets must be at least 1");
  let i = n.readSync(a.dataId), o = PX(i, r);
  return n.makeTensorInfo(a.shape, "int32", o);
}
var lte = { kernelName: Pg, backendName: "webgl", kernelFunc: ute };
var cte = "return tan(x);";
var dte = Ke({ opSnippet: cte });
var pte = { kernelName: Ho, backendName: "webgl", kernelFunc: dte };
var hte = `
  float e2x = exp(-2.0 * abs(x));
  return sign(x) * (1.0 - e2x) / (1.0 + e2x);
`;
var fte = Ke({ opSnippet: hte });
var mte = { kernelName: fi, backendName: "webgl", kernelFunc: fte };
var gte = class {
  constructor(e, t) {
    this.variableNames = ["A"];
    let n = new Array(e.length);
    for (let a = 0; a < n.length; a++)
      n[a] = e[a] * t[a];
    this.outputShape = n, this.rank = n.length;
    let s = ot(this.rank), r = bte(e);
    this.userCode = `
      void main() {
        ${s} resRC = getOutputCoords();
        setOutput(getA(${r}));
      }
    `;
  }
};
function bte(e) {
  let t = e.length;
  if (t > 5)
    throw Error(`Tile for rank ${t} is not yet supported`);
  if (t === 1)
    return `imod(resRC, ${e[0]})`;
  let n = ["resRC.x", "resRC.y", "resRC.z", "resRC.w", "resRC.u"], s = [];
  for (let r = 0; r < e.length; r++)
    s.push(`imod(${n[r]}, ${e[r]})`);
  return s.join();
}
function I2(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { reps: a } = s;
  if (r.dtype === "string" || r.shape.length > 5) {
    let u = n.readSync(r.dataId), l = r.dtype === "string" ? u.map((d) => w.decodeString(d)) : u, c = Ae(r.shape, r.dtype, l), p = MX(c, a);
    return n.makeTensorInfo(p.shape, p.dtype, p.values);
  }
  let i = new gte(r.shape, a);
  return n.runWebGLProgram(i, [r], r.dtype);
}
var yte = { kernelName: Cr, backendName: "webgl", kernelFunc: I2 };
var vte = class {
  constructor(e) {
    this.variableNames = ["x", "indices"], this.customUniforms = [{ name: "n", type: "int" }, { name: "firstPass", type: "int" }, { name: "negativeInf", type: "float" }, { name: "dir", type: "int" }, { name: "inc", type: "int" }], this.outputShape = e, this.userCode = `
       void main() {
         ivec2 coords = getOutputCoords();
         int batch = coords[0];
         int elemIdx = coords[1];

         // We compare elements pair-wise within a group of size 2 * inc.
         // The comparing rule for each group alternates between ascending
         // and descending. Within each group, we compare each pair at
         // positions i and i+inc. To decide whether an element at position i
         // is x0 or x1, we mod it by 2 * inc, if the result is smaller than
         // inc, it is in the first half of the group, we denote it as x0,
         // otherwise we denote it as x1.
         // For example, as shown in the Bitonic top K paper referenced above,
         // Figure5(a) shows that element[1] is in the
         // second half of the group when group size is 2, but it is in the
         // first half of the group when group size is 4.

         bool isFirstInPair = imod(elemIdx, 2 * inc) < inc;
         int i = isFirstInPair ? elemIdx : elemIdx - inc;

         int i0 = firstPass == 1 ? i : int(getIndices(batch, i));
         int i1 = firstPass == 1 ? i + inc : int(getIndices(batch, i + inc));
         float x0 = i0 < n ? getX(batch, i0) : negativeInf;
         float x1 = i1 < n ? getX(batch, i1) : negativeInf;

         // Denotes which direction indices are in (ascending or descending).
         bool reverse = imod(elemIdx, 2 * dir) >= dir;
         bool isGreater = x0 > x1 || (x0 == x1 && i1 > i0);
         if (reverse == isGreater) { // Elements in opposite order of direction
           int iTemp = i0;
           i0 = i1;
           i1 = iTemp;
         }
         if (isFirstInPair) {
            setOutput(float(i0));
         } else {
            setOutput(float(i1));
         }
       }
     `;
  }
};
var xte = class {
  constructor(e) {
    this.variableNames = ["x", "indices"], this.customUniforms = [{ name: "n", type: "int" }, { name: "firstPass", type: "int" }, { name: "k", type: "int" }], this.outputShape = e, this.userCode = `
    void main() {
         // Takes max of indices (0, k), (1, k + 1), (2, k + 2) ...
         ivec2 coords = getOutputCoords();
         int batch = coords[0];
         int elemIdx = coords[1];

         // The output size is half of the previous size.
         // If the previous sequence is | | | | _ _ _ _  | | | |  _ _ _ _ (k=4),
         // we only need to output the indices at positions |, the indices at
         // positions _ can be thrown away, see Figure5(b) After Phase 2
         // (Merge phase) in the Bitonic Top K paper referenced above.
         // For example, the paper shows we only need to output the orange bars.
         // The output sequence should look like this | | | | | | | |.
         // Because the sequence is halved, to map the output index back
         // to the previous sequence to find the corresponding value,
         // we need to double the index. When we double the index,
         // we basically interpolate a position, so 2i looks like
         // | _ | _ | _ | _ | _ | _ | _. We move the | to the first k position
         // of each 2k positions by - elemIdx % k. E.g. for output at
         // index 4,5,6,7, we want to get the corresponding element at
         // original index 8,9,10,11, for output at index 8,9,10,11,
         // we want to get the corresponding element at original index
         // 16,17,18,19, so on and so forth.

         int i = elemIdx < k ? elemIdx : (elemIdx * 2 - imod(elemIdx, k));
         int i0 = firstPass == 1 ? i : int(getIndices(batch, i));
         int i1 = firstPass == 1 ? i + k : int(getIndices(batch, i + k));

         float x0 = getX(batch, i0);
         float x1 = i1 < n ? getX(batch, i1) : x0;

         setOutput(x0 >= x1 ? float(i0) : float(i1));
       }
     `;
  }
};
function Gr(e, t) {
  t !== null && e.disposeIntermediateTensorInfo(t);
}
function Rw(e) {
  let t = 1;
  for (; t < e; )
    t *= 2;
  return t;
}
function wte(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { k: a, sorted: i } = s, o = K().getNumber("TOPK_LAST_DIM_CPU_HANDOFF_SIZE_THRESHOLD"), u = K().getNumber("TOPK_K_CPU_HANDOFF_THRESHOLD"), l = r.shape, c = l[l.length - 1];
  if (n.shouldExecuteOnCPU([r]) || c < o || a > u) {
    let P = n.readSync(r.dataId), [A, O] = LX(P, l, r.dtype, a, i);
    return [n.makeTensorInfo(A.shape, A.dtype, A.values), n.makeTensorInfo(O.shape, O.dtype, O.values)];
  }
  if (a === 0)
    return l[l.length - 1] = 0, [n.makeTensorInfo(l, r.dtype, []), n.makeTensorInfo(l, "int32", [])];
  if (c === 1)
    return [r, nc({ attrs: { shape: l, dtype: "int32", value: 0 }, backend: n })];
  let p = n.texData.get(r.dataId), d = p !== null && p.isPacked, h = d ? n.unpackTensor(r) : r, m = w.sizeFromShape(l) / c, g = he({ inputs: { x: h }, attrs: { shape: [m, c] }, backend: n });
  d && Gr(n, h);
  let b = Rw(a), y = Rw(c), v = null, x = () => v === null ? [g, g] : [g, v], k = (P, A, O) => {
    let T = x(), z = new vte(O), q = [[c], [v === null ? 1 : 0], [Number.NEGATIVE_INFINITY], [P], [A]], X = v;
    v = n.runWebGLProgram(z, T, "int32", q), Gr(n, X);
  };
  for (let P = 1; P < b; P *= 2) {
    let A = P * 2;
    for (let O = P; O >= 1; O /= 2)
      k(A, O, [m, y]);
  }
  for (let P = y; P > b; P /= 2) {
    let A = x(), O = new xte([m, P / 2]), z = [[c], [v === null ? 1 : 0], [b]], W = v;
    v = n.runWebGLProgram(O, A, "int32", z), Gr(n, W);
    let q = b / 2, X = q * 2;
    for (let Y = q; Y >= 1; Y /= 2)
      k(X, Y, v.shape);
  }
  let I = v;
  v = pu({ inputs: { x: v }, backend: n, attrs: { begin: 0, size: [m, a] } }), Gr(n, I);
  let $ = m2({ inputs: { x: g, indices: v }, backend: n, attrs: { axis: 1, batchDims: 1 } });
  Gr(n, g);
  let R = l.slice(0, -1);
  R.push(a), I = v, v = he({ inputs: { x: v }, attrs: { shape: R }, backend: n }), Gr(n, I);
  let E = $;
  return $ = he({ inputs: { x: $ }, attrs: { shape: R }, backend: n }), Gr(n, E), [$, v];
}
var kte = { kernelName: qo, backendName: "webgl", kernelFunc: wte };
var Ste = class {
  constructor(e, t, n, s, r, a) {
    this.variableNames = ["Image", "Transforms"], this.outputShape = a;
    let i = n === "nearest" ? 1 : 2, o;
    switch (s) {
      case "constant":
        o = 1;
        break;
      case "reflect":
        o = 2;
        break;
      case "wrap":
        o = 3;
        break;
      case "nearest":
        o = 4;
        break;
      default:
        o = 1;
        break;
    }
    this.userCode = `
            float mapCoord(float outCoord, float len) {
              float inCoord = outCoord;
              if(${o} == 2) {
                if (inCoord < 0.0) {
                  if (len <= 1.0) {
                    inCoord = 0.0;
                  } else {
                    float sz2 = 2.0 * len;
                    if (inCoord < sz2) {
                      inCoord = sz2 * float(int(float(-inCoord / sz2))) +
                      inCoord;
                    }
                    inCoord = inCoord < -len ? inCoord + sz2 : -inCoord - 1.0;
                  }
                } else if (inCoord > len - 1.0) {
                  if (len <= 1.0) {
                    inCoord = 0.0;
                  } else {
                    float sz2 = 2.0 * len;
                    inCoord -= sz2 * float(int(float(inCoord / sz2)));
                    if (inCoord >= len) {
                      inCoord = sz2 - inCoord - 1.0;
                    }
                  }
                }
                return clamp(inCoord, 0.0, len - 1.0);
              } else if (${o} == 3) {
                if (inCoord < 0.0) {
                  if (len <= 1.0) {
                    inCoord = 0.0;
                  } else {
                    float sz = len - 1.0;
                    inCoord += len * (float(int(float(-inCoord / sz))) + 1.0);
                  }
                } else if (inCoord > len - 1.0) {
                  if (len <= 1.0) {
                    inCoord = 0.0;
                  } else {
                    float sz = len - 1.0;
                    inCoord -= len * float(int(float(inCoord / sz)));
                  }
                }
                return clamp(inCoord, 0.0, len - 1.0);
              } else if (${o} == 4) {
                return clamp(outCoord, 0.0, len - 1.0);
              } else {
                return outCoord;
              }
            }

            float readWithFillValue(int batch, int coordY, int coordX,
              int channel) {
              float outputValue;
              if (0 <= coordY && coordY < ${e} && 0 <= coordX && coordX < ${t}) {
                  outputValue = getImage(batch, coordY, coordX, channel);
              } else {
                outputValue = float(${r});
              }
              return outputValue;
            }

            void main() {
              ivec4 coords = getOutputCoords();
              float outputValue;
              int batch = coords[0];
              int x = coords[2];
              int y = coords[1];
              int channel = coords[3];
              float xf = float(x);
              float yf = float(y);
              float a1 = getTransforms(batch, 0);
              float a2 = getTransforms(batch, 1);
              float a3 = getTransforms(batch, 2);
              float b1 = getTransforms(batch, 3);
              float b2 = getTransforms(batch, 4);
              float b3 = getTransforms(batch, 5);
              float c1 = getTransforms(batch, 6);
              float c2 = getTransforms(batch, 7);
              float projection = c1 * xf + c2 * yf + 1.0;
              if (projection == 0.0) {
                outputValue = float(${r});
              } else {
                float inX = (a1 * xf + a2 * yf + a3) / projection;
                float inY = (b1 * xf + b2 * yf + b3) / projection;
                float mapX = mapCoord(inX, float(${t}));
                float mapY = mapCoord(inY, float(${e}));

                if (${i} == 1) {
                  int coordY = int(round(mapY));
                  int coordX = int(round(mapX));
                  outputValue = readWithFillValue(batch, coordY, coordX,
                    channel);
                } else {
                  float yFloor = floor(mapY);
                  float xFloor = floor(mapX);
                  float yCeil = yFloor + 1.0;
                  float xCeil = xFloor + 1.0;
                  float valueYFloor = (xCeil - mapX) *
                  readWithFillValue(batch, int(yFloor), int(xFloor), channel) +
                  (mapX - xFloor) *
                  readWithFillValue(batch, int(yFloor), int(xCeil), channel);
                  float valueYCeil = (xCeil - mapX) *
                  readWithFillValue(batch, int(yCeil), int(xFloor), channel) +
                  (mapX - xFloor) *
                  readWithFillValue(batch, int(yCeil), int(xCeil), channel);
                  outputValue = (yCeil - mapY) * valueYFloor +
                  (mapY - yFloor) * valueYCeil;
                }
              }
              setOutput(outputValue);
            }
        `;
  }
};
function Ite(e) {
  let { inputs: t, backend: n, attrs: s } = e, { image: r, transforms: a } = t, { interpolation: i, fillMode: o, fillValue: u, outputShape: l } = s, [c, p, d, h] = r.shape, [f, m] = l != null ? l : [p, d], g = [c, f, m, h], b = new Ste(p, d, i, o, u, g);
  return n.runWebGLProgram(b, [r, a], "float32");
}
var Cte = { kernelName: jo, backendName: "webgl", kernelFunc: Ite };
function Nte(e) {
  let { inputs: t, attrs: n, backend: s } = e, { axis: r } = n, { x: a } = t;
  iu(a, "unique"), console.warn("WARNING: ", "UI might be locked temporarily as data is being downloaded");
  let i = s.readSync(a.dataId), { outputValues: o, outputShape: u, indices: l } = BX(i, r, a.shape, a.dtype);
  return [s.makeTensorInfo(u, a.dtype, o), s.makeTensorInfo([l.length], "int32", l)];
}
var Tte = { kernelName: zg, backendName: "webgl", kernelFunc: Nte };
function $te(e) {
  let { inputs: t, backend: n, attrs: s } = e, { value: r } = t, { axis: a } = s;
  a < 0 && (a += r.shape.length);
  let i = r, o = i.shape.length, u = r.shape[a], l = new Array(o - 1), c = 0;
  for (let m = 0; m < o; m++)
    m !== a && (l[c++] = i.shape[m]);
  let p = [], d = new Array(o).fill(0), h = i.shape.slice();
  h[a] = 1;
  let f = new Array(u);
  for (let m = 0; m < f.length; m++) {
    d[a] = m;
    let g = pu({ inputs: { x: i }, backend: n, attrs: { begin: d, size: h } }), b = he({ inputs: { x: g }, backend: n, attrs: { shape: l } });
    f[m] = b, p.push(g);
  }
  return p.forEach((m) => n.disposeIntermediateTensorInfo(m)), f;
}
var _te = { kernelName: Ko, backendName: "webgl", kernelFunc: $te };
var Ate = class {
  constructor(e, t) {
    this.variableNames = ["x", "segmentIds"];
    let n = e.windowSize, s = e.batchSize, r = e.inSize, a = e.numSegments, i = a * Math.ceil(r / n);
    this.outputShape = [s, i];
    let o = "0.0", u = "sumValue", l = Math.floor(n / 4) * 4, c = n % 4, p = `
        sumValue += dot(values, segFilter);
    `, d = "";
    r % n > 0 && (d = `
        if (inIdx < 0 || inIdx >= ${r}) {
          return initializationValue;
        }
      `);
    let h = "";
    r % n > 0 && (h = `
        if (inIdx < 0 || inIdx >= ${r}) {
          return -1.0;
        }
      `), this.userCode = `
      const float initializationValue = ${o};

      float getValue(int batch, int inIdx) {
        ${d}
        return getX(batch, inIdx);
      }

      float getSegmentIdAtIndex(int inIdx) {
        ${h}
        return getSegmentIds(inIdx);
      }

      void main() {
        ivec2 coords = getOutputCoords();
        int batch = coords[0];
        int outIdx = coords[1];
        int inOffset = int(floor(float(outIdx) / float(
          ${a})) * float(${n}));
        int currentSeg = int(mod(float(outIdx), float(${a})));

        float sumValue = 0.0;

        for (int i = 0; i < ${l}; i += 4) {
          int inIdx = inOffset + i;
          vec4 values = vec4(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1),
            getValue(batch, inIdx + 2),
            getValue(batch, inIdx + 3)
          );

          vec4 segFilter = vec4(
            int(getSegmentIdAtIndex(inIdx)) == currentSeg ? 1 : 0,
            int(getSegmentIdAtIndex(inIdx + 1)) == currentSeg ? 1 : 0,
            int(getSegmentIdAtIndex(inIdx + 2)) == currentSeg ? 1 : 0,
            int(getSegmentIdAtIndex(inIdx + 3)) == currentSeg ? 1 : 0
          );

          ${p}
        }

        int inIdx = inOffset + ${l};
        if (${c === 1}) {
          vec4 values = vec4(
            getValue(batch, inIdx),
            initializationValue,
            initializationValue,
            initializationValue
          );

          int inIdxSeg = int(getSegmentIdAtIndex(inIdx));

          vec4 segFilter = vec4(
            int(getSegmentIdAtIndex(inIdx)) == currentSeg ? 1 : 0,
            0,
            0,
            0
          );

          ${p}
        } else if (${c === 2}) {
          vec4 values = vec4(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1),
            initializationValue,
            initializationValue
          );

          vec4 segFilter = vec4(
            int(getSegmentIdAtIndex(inIdx)) == currentSeg ? 1 : 0,
            int(getSegmentIdAtIndex(inIdx + 1)) == currentSeg ? 1 : 0,
              0,
              0
          );

          ${p}
        } else if (${c === 3}) {
          vec4 values = vec4(
            getValue(batch, inIdx),
            getValue(batch, inIdx + 1),
            getValue(batch, inIdx + 2),
            initializationValue
          );

          vec4 segFilter = vec4(
            int(getSegmentIdAtIndex(inIdx)) == currentSeg ? 1 : 0,
            int(getSegmentIdAtIndex(inIdx + 1)) == currentSeg ? 1 : 0,
            int(getSegmentIdAtIndex(inIdx + 2)) == currentSeg ? 1 : 0,
            0
          );

          ${p}
        }
        setOutput(${u});
      }
    `;
  }
};
function Ete(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, segmentIds: a } = t, { numSegments: i } = s, o = r.shape.length, u = [], l = 0, c = C.getAxesPermutation([l], o), p = r;
  c != null && (p = _t({ inputs: { x: r }, backend: n, attrs: { perm: c } }), u.push(p), l = C.getInnerMostAxes(1, o)[0]);
  let d = C.segment_util.computeOutShape(p.shape, l, i), h = w.sizeFromShape([p.shape[l]]), f = he({ inputs: { x: p }, backend: n, attrs: { shape: [-1, h] } });
  u.push(f);
  let m = bp(r.dtype), g = (x, k, I, $, R) => {
    let E = x.shape[0], P = x.shape[1], A = C.segment_util.segOpComputeOptimalWindowSize(P, R), O = { windowSize: A, inSize: P, batchSize: E, numSegments: R }, T = new Ate(O, k), z = n.compileAndRun(T, [x, I], $);
    if (u.push(z), z.shape[1] === R)
      return z;
    let W = k2({ backend: n, attrs: { start: 0, stop: R, step: 1, dtype: "float32" } }), q = I2({ inputs: { x: W }, backend: n, attrs: { reps: [P / A] } });
    return u.push(W), u.push(q), g(z, k, q, $, R);
  }, b = g(f, "unsortedSegmentSum", a, m, i), y = he({ inputs: { x: b }, backend: n, attrs: { shape: d } }), v = y;
  if (c != null) {
    u.push(y);
    let x = C.getUndoAxesPermutation(c);
    v = _t({ inputs: { x: v }, backend: n, attrs: { perm: x } });
  }
  return u.forEach((x) => n.disposeIntermediateTensorInfo(x)), v;
}
var Rte = { kernelName: mp, backendName: "webgl", kernelFunc: Ete };
var Dte = [F8, P8, L8, W8, G8, j8, X8, Q8, tY, sY, iY, lY, pY, gY, vY, wY, SY, TY, _Y, EY, OY, WY, GY, qY, ZY, e9, r9, m8, o9, p9, g9, k9, I9, N9, $9, A9, D9, P9, L9, V9, U9, H9, K9, Y9, eQ, nQ, aQ, uQ, cQ, fQ, yQ, kQ, CQ, $Q, _Q, EQ, DQ, OQ, zQ, LQ, UQ, qQ, XQ, QQ, eZ, sZ, oZ, dZ, f8, hZ, c9, gZ, vZ, kZ, b8, NZ, AZ, RZ, PZ, LZ, UZ, qZ, YZ, e7, s7, a7, l7, d7, h7, b7, v7, w7, S7, C7, _7, D7, z7, H7, k8, X7, Z7, tJ, rJ, KY, oJ, lJ, dJ, fJ, yJ, v8, xJ, wJ, XY, V7, IJ, $J, RJ, I8, PJ, LJ, UJ, qJ, YJ, ZJ, tee, ree, iee, lee, pee, mee, vee, kee, Cee, $ee, BY, U7, Eee, Dee, Oee, zee, Lee, Vee, Uee, Hee, jee, Yee, Zee, ete, ste, ate, ote, lte, W7, E8, pte, mte, yte, kte, Cte, R8, Tte, _te, Rte, uJ];
for (let e of Dte)
  Fl(e);
var Dr = K();
Dr.registerFlag("WEBGPU_DEFERRED_SUBMIT_BATCH_SIZE", () => 15);
Dr.registerFlag("WEBGPU_CPU_FORWARD", () => true);
Dr.registerFlag("WEBGPU_MATMUL_WORK_PER_THREAD", () => 4);
Dr.registerFlag("WEBGPU_USE_NAIVE_CONV2D_TRANSPOSE", () => false);
Dr.registerFlag("WEBGPU_USE_LOW_POWER_GPU", () => false);
Dr.registerFlag("WEBGPU_CPU_HANDOFF_SIZE_THRESHOLD", () => 1e3);
Dr.registerFlag("WEBGPU_USE_PROFILE_TOOL", () => false);
Dr.registerFlag("WEBGPU_USE_IMPORT", () => false);
var Fte = "return a + b;";
var Ote = "return areal * breal - aimag * bimag;";
var Pte = "return areal * bimag + aimag * breal;";
var zte = "return a / b;";
var Mte = "return a * b;";
var Lte = "return (a - b) * (a - b);";
var Bte = "return a - b;";
var Vte = "return f32(a == b);";
var Wte = "return vec4<f32>(a == b);";
var Ute = "return f32(a > b);";
var Gte = "return vec4<f32>(a > b);";
var Hte = "return f32(a >= b);";
var qte = "return vec4<f32>(a >= b);";
var jte = "return f32(a < b);";
var Kte = "return vec4<f32>(a < b);";
var Xte = "return f32(a <= b);";
var Yte = "return vec4<f32>(a <= b);";
var Qte = "return f32(f32(a) >= 1.0 && f32(b) >= 1.0);";
var Zte = `return (vec4<f32>(a >= vec4<f32>(1.0)) *
  vec4<f32>(b >= vec4<f32>(1.0)));`;
var Jte = `
  if (isnan(a)) { return a; }
  if (isnan(b)) { return b; }
  `;
var C2 = `
  if (isNaN.r) {
    resultTemp.r = uniforms.NAN;
  }
  if (isNaN.g) {
    resultTemp.g = uniforms.NAN;
  }
  if (isNaN.b) {
    resultTemp.b = uniforms.NAN;
  }
  if (isNaN.a) {
    resultTemp.a = uniforms.NAN;
  }
  `;
var ene = `
  let s = sign(a) * sign(b);
  let ia = i32(round(a));
  let ib = i32(round(b));
  return f32(idiv(ia, ib, s));
  `;
var tne = `
  let ia = vec4<i32>(round(a));
  let ib = vec4<i32>(round(b));
  let cond = ib != vec4<i32>(0);
  var resultTemp = vec4<i32>(0);
  let s = sign(a) * sign(b);

  // Windows (D3D) wants guaranteed non-zero int division at compile-time.
  if (cond[0]) {
    resultTemp[0] = idiv(ia[0], ib[0], s[0]);
  }
  if (cond[1]) {
    resultTemp[1] = idiv(ia[1], ib[1], s[1]);
  }
  if (cond[2]) {
    resultTemp[2] = idiv(ia[2], ib[2], s[2]);
  }
  if (cond[3]) {
    resultTemp[3] = idiv(ia[3], ib[3], s[3]);
  }
  return vec4<f32>(resultTemp);
  `;
var nne = "return f32(a != b);";
var sne = "return vec4<f32>(a != b);";
var rne = `
  if(a < 0.0 && floor(b) < b) {
    return uniforms.NAN;
  }
  if (b == 0.0) {
    return 1.0;
  }
  if (round(abs(b) % 2.0) != 1.0) {
    return pow(abs(a), b);
  }
  return sign(a) * pow(abs(a), b);
  `;
var ane = `
  let isModRound1Bool = vec4<i32>(round(abs(b) % vec4<f32>(2.0))) == vec4<i32>(1);
  let isModRound1 = vec4<f32>(isModRound1Bool);
  let multiplier = sign(a) * isModRound1 + (vec4<f32>(1.0) - isModRound1);
  var resultTemp = multiplier * pow(abs(a), b);

  // Ensure that a^0 = 1, including 0^0 = 1 as this correspond to TF and JS
  let isExpZero = b == vec4<f32>(0.0);
  if (isExpZero.r) {
    resultTemp.r = 1.0;
  }
  if (isExpZero.g) {
    resultTemp.g = 1.0;
  }
  if (isExpZero.b) {
    resultTemp.b = 1.0;
  }
  if (isExpZero.a) {
    resultTemp.a = 1.0;
  }
  let isNaN = a < vec4<f32>(0.0) & floor(b) < b;
  ${C2}
  return resultTemp;
  `;
var ine = "if (a < 0.0) { return b * a; }  return a;";
var one = `
  let aLessThanZero = vec4<f32>(a < vec4<f32>(0.0));
  return (aLessThanZero * (b * a)) + ((vec4<f32>(1.0) - aLessThanZero) * a);
  `;
function Dw(e, t) {
  let n = t ? C2 : Jte;
  return t ? `
    var resultTemp = vec4<f32>(${e}(a, b));
    let isNaN = isnanVec4(a) | isnanVec4(b);
    ` + n + `
    return resultTemp;
  ` : n + `
    return ${e}(a, b);
  `;
}
function sc(e, t) {
  switch (e) {
    case 0:
      return Mte;
    case 1:
      return Fte;
    case 2:
      return Bte;
    case 3:
      return zte;
    case 4:
      return t ? Wte : Vte;
    case 5:
      return t ? Gte : Ute;
    case 6:
      return t ? qte : Hte;
    case 7:
      return t ? Kte : jte;
    case 8:
      return t ? Yte : Xte;
    case 9:
      return t ? Zte : Qte;
    case 10:
      return t ? sne : nne;
    case 11:
      return Lte;
    case 12:
      return t ? tne : ene;
    case 14:
      return t ? one : ine;
    case 15:
      return Dw("max", t);
    case 16:
      return Dw("min", t);
    case 13:
      return t ? ane : rne;
    case 17:
      return Ote;
    case 18:
      return Pte;
    default:
      throw new Error(`BinaryType ${e} is not implemented!`);
  }
}
var une = "return abs(a);";
var lne = "return ceil(a);";
var cne = "return cos(a);";
var dne = `
  let e2x = exp(-a);
  return (e2x + 1.0 / e2x) / 2.0;
`;
var pne = "return exp(a) - 1.0;";
var hne = "if (a >= 0.0) { return a; }  return (exp(a) - 1.0);";
var fne = `
  var resFloat = exp(a) - vec4<f32>(1.0);
  if (a.r >= 0.0) {
    resFloat.r = a.r;
  }
  if (a.g >= 0.0) {
    resFloat.g = a.g;
  }
  if (a.b >= 0.0) {
    resFloat.b = a.b;
  }
  if (a.a >= 0.0) {
    resFloat.a = a.a;
  }
  return resFloat;
`;
var mne = "return exp(a);";
var gne = "return floor(a);";
var bne = "return a;";
var yne = `if (a < 0.0) { return 1.0/0.0; }
  return log(a);`;
var vne = "return f32(!(a >= 1.0));";
var xne = "return -a;";
var wne = "if (a < 0.0) { return uniforms.alpha * a; } return a;";
var kne = `
  let aLessThanZero = vec4<f32>(a < vec4<f32>(0.0));
  return (aLessThanZero * (uniforms.alpha * a)) + ((vec4<f32>(1.0) - aLessThanZero) * a);
`;
var Sne = "return select(a, 0.0, a < 0.0);";
var Ine = "return clamp(a, 0.0, 6.0);";
var Cne = "return clamp(a, vec4<f32>(0.0, 0.0, 0.0, 0.0), vec4<f32>(6.0, 6.0, 6.0, 6.0));";
var Nne = `
  return select(a, vec4<f32>(0.0), a < vec4<f32>(0.0));
`;
var Tne = "return 1.0/sqrt(a);";
var $ne = "return 1.0 / (1.0 + exp(-1.0 * a));";
var _ne = "return sin(a);";
var Ane = `
  let e2x = exp(a);
  return (e2x - 1.0 / e2x) / 2.0;
`;
var Ene = "return sqrt(a);";
var Rne = "return a * a;";
var Dne = `
  let e2x = exp(-2.0 * abs(a));
  return sign(a) * (1.0 - e2x) / (1.0 + e2x);
`;
var Fne = "return f32(i32((a)));";
function qr(e, t) {
  switch (e) {
    case 0:
      return une;
    case 2:
      return cne;
    case 3:
      return dne;
    case 1:
      return lne;
    case 4:
      return t ? fne : hne;
    case 5:
      return mne;
    case 6:
      return pne;
    case 7:
      return gne;
    case 8:
      return bne;
    case 9:
      return yne;
    case 10:
      return vne;
    case 11:
      return xne;
    case 14:
      return t ? kne : wne;
    case 12:
      return t ? Nne : Sne;
    case 13:
      return t ? Cne : Ine;
    case 15:
      return Tne;
    case 18:
      return $ne;
    case 16:
      return _ne;
    case 17:
      return Ane;
    case 19:
      return Ene;
    case 20:
      return Rne;
    case 21:
      return Dne;
    case 22:
      return Fne;
    default:
      throw new Error(`BinaryType ${e} is not implemented!`);
  }
}
function Fr(e, t = false) {
  if (e === null)
    return null;
  if (e === "linear")
    return qr(8);
  if (e === "relu")
    return qr(12, t);
  if (e === "elu")
    return qr(4, t);
  if (e === "relu6")
    return qr(13, t);
  if (e === "prelu")
    return sc(14, t);
  if (e === "sigmoid")
    return qr(18, t);
  if (e === "leakyrelu")
    return qr(14, t);
  throw new Error(`Activation ${e} has not been implemented for the WebGPU backend.`);
}
function One(e, t) {
  if (Math.max(...e) > 3)
    throw new Error("Cannot symbolically compute strides for rank > 4 tensor.");
  let n = e.length, s = e.map((a) => `${t}[${a}]`), r = new Array(n - 1);
  r[n - 2] = s[n - 1];
  for (let a = n - 3; a >= 0; --a)
    r[a] = `(${r[a + 1]} * ${s[a + 1]})`;
  return r;
}
function Ut(e) {
  if (e <= 1)
    return "i32";
  if (e === 2)
    return "vec2<i32>";
  if (e === 3)
    return "vec3<i32>";
  if (e === 4)
    return "vec4<i32>";
  if (e === 5)
    return "vec5";
  if (e === 6)
    return "vec6";
  throw Error(`GPU for rank ${e} is not yet supported`);
}
function pr(e) {
  if (e === 0)
    return "x";
  if (e === 1)
    return "y";
  if (e === 2)
    return "z";
  if (e === 3)
    return "w";
  if (e === 4)
    return "u";
  if (e === 5)
    return "v";
  throw Error(`Index ${e} is not yet supported`);
}
function hd(e, t) {
  return e === "float32" ? t ? "vec4<f32>" : "f32" : e === "int32" || e === "bool" ? t ? "vec4<i32>" : "i32" : e;
}
function Ev() {
  return `
  @stage(compute) @workgroup_size(workGroupSizeX, workGroupSizeY, workGroupSizeZ)
`;
}
function Ii() {
  return `
  ${Ev()}
  fn main(@builtin(local_invocation_id) LocalId : vec3<u32>,
          @builtin(global_invocation_id) GlobalId : vec3<u32>,
          @builtin(num_workgroups) NumWorkgroups: vec3<u32>) {
    localId = LocalId;
    globalId = GlobalId;
    numWorkgroups = NumWorkgroups;
`;
}
function Ue() {
  return `
    ${Ii()}
      let index = getGlobalIndex();
`;
}
function Pne(e, t, n, s = false) {
  let r = [];
  if (r.push(`
    let workGroupSizeX = ${n.workGroupSize[0]}u;
    let workGroupSizeY = ${n.workGroupSize[1]}u;
    let workGroupSizeZ = ${n.workGroupSize[2]}u;

    var<private> localId: vec3<u32>;
    var<private> globalId: vec3<u32>;
    var<private> numWorkgroups: vec3<u32>;

    // Only used when the y/z dimension of workgroup size is 1.
    fn getGlobalIndex() -> i32 {
      if (numWorkgroups.y == 1u && numWorkgroups.z == 1u) {
        return i32(globalId.x);
      }

      let localInvocationIndex = localId.z * workGroupSizeX * workGroupSizeY +
          localId.y * workGroupSizeX + localId.x;
      let workGroupID = (globalId - localId)/vec3<u32>(
          workGroupSizeX, workGroupSizeY, workGroupSizeZ);

      return i32((workGroupID.z * numWorkgroups.x * numWorkgroups.y +
        workGroupID.y * numWorkgroups.x + workGroupID.x) *
        (workGroupSizeX * workGroupSizeY * workGroupSizeZ) +
        localInvocationIndex);
    }
  `), s === true)
    return r.push(`
      struct Uniform {
        size            : i32,
        numChannels     : i32,
        outShapeStrides : vec2<i32>,
        dispatchSize    : vec3<u32>,
      };

      @group(0) @binding(0) var<storage, write> result: array<${hd(t.dtype, n.isVec4)}>;
      @group(0) @binding(2) var<uniform> uniforms: Uniform;
    `), [Fw, r.join(`
`), Ow(t.shape), n.getUserCode()].join(`
`);
  let a = false, i = false, o = "struct Uniforms { NAN : f32, ";
  n.variableNames.forEach((m, g) => {
    let b = Ut(e[g].shape.length);
    (b === "vec5" || b === "vec6") && (i = true), (a || i) && (o += "@align(16) "), a = i, o += `${m.charAt(0).toLowerCase() + m.slice(1)}Shape : ${b}, `;
  });
  let u = Ut(t.shape.length);
  i = u === "vec5" || u === "vec6", (a || i) && (o += "@align(16) "), a = i, o += `outShape : ${u}, `;
  let l = t.shape.length - 1, c = Ut(l);
  i = c === "vec5" || c === "vec6", (a || i) && (o += "@align(16) "), a = i, o += `
       outShapeStrides: ${c}, `, n.size && (a && (o += "@align(16) "), a = false, o += "size : i32, "), n.uniforms && (a && (o += "@align(16) "), o += n.uniforms), o += "};", r.push(o), n.atomic ? r.push(`
    @group(0) @binding(0) var<storage, read_write> result: array<atomic<i32>>;
  `) : r.push(`
    @group(0) @binding(0) var<storage, write> result: array<${hd(t.dtype, n.isVec4)}>;
  `), n.variableNames.forEach((m, g) => {
    r.push(`
    @group(0) @binding(${1 + g}) var<storage, read> ${m}: array<${hd(e[g].dtype, n.isVec4)}>;
    `);
  }), o !== "" && r.push(`
    @group(0) @binding(${1 + n.variableNames.length}) var<uniform> uniforms: Uniforms;
    `);
  let [p, d] = Wne(t.shape, n.dispatchLayout), h = [Fw, r.join(`
`), Ow(t.shape), p, zne(t.shape.length)];
  if (n.atomic || h.push(Mne(t.shape, t.dtype, n.isVec4)), d === t.shape.length) {
    let m = e.map((g) => Lne(g, t.shape, n.isVec4, n.dispatchLayout.x.length === t.shape.length)).join(`
`);
    h.push(m);
  }
  return h.push(n.getUserCode()), h.join(`
`);
}
var Fw = `
  struct vec5 {x: i32, y: i32, z: i32, w: i32, u: i32};
  struct vec6 {x: i32, y: i32, z: i32, w: i32, u: i32, v: i32};

  // Checks whether coordinates lie within the bounds of the shape.
  fn coordsInBounds2D(coord : vec2<i32>, shape : vec2<i32>) -> bool {
    return all(coord >= vec2<i32>(0)) && all(coord < shape);
  }
  fn coordsInBounds3D(coord : vec3<i32>, shape : vec3<i32>) -> bool {
    return all(coord >= vec3<i32>(0)) && all(coord < shape);
  }
  fn coordsInBounds4D(coord : vec4<i32>, shape : vec4<i32>) -> bool {
    return all(coord >= vec4<i32>(0)) && all(coord < shape);
  }

  fn getIndexFromCoords1D(coord : i32, shape : i32) -> i32 {
    return coord;
  }
  fn getIndexFromCoords2D(coords : vec2<i32>, shape : vec2<i32>) -> i32 {
    return dot(coords, vec2<i32>(shape.y, 1));
  }
  fn getIndexFromCoords3D(coords : vec3<i32>, shape : vec3<i32>) -> i32 {
    return dot(coords, vec3<i32>(shape.y * shape.z, shape.z, 1));
  }
  fn getIndexFromCoords4D(coords : vec4<i32>, shape : vec4<i32>) -> i32 {
    return dot(coords, vec4<i32>(
        shape.y * shape.z * shape.w, shape.z * shape.w, shape.w, 1));
  }
  fn getIndexFromCoords5D(coords : vec5, shape : vec5) -> i32 {
    let shapeStrides: vec5 = vec5(shape.y * shape.z * shape.w * shape.u, shape.z * shape.w * shape.u, shape.w * shape.u, shape.u, 1);
    return coords.x*shapeStrides.x + coords.y*shapeStrides.y + coords.z*shapeStrides.z + coords.w*shapeStrides.w + coords.u*shapeStrides.u;
  }
  fn getIndexFromCoords6D(coords : vec6, shape : vec6) -> i32 {
    let shapeStrides: vec6 = vec6(shape.y * shape.z * shape.w * shape.u * shape.v, shape.z * shape.w * shape.u * shape.v, shape.w * shape.u * shape.v, shape.u * shape.v, shape.v, 1);
    return coords.x*shapeStrides.x + coords.y*shapeStrides.y + coords.z*shapeStrides.z + coords.w*shapeStrides.w + coords.u*shapeStrides.u + coords.v*shapeStrides.v;
  }

  fn idiv(a: i32, b: i32, sign: f32) -> i32 {
    var res: i32 = a / b;
    let mod: i32 = a % b;
    if (sign < 0. && mod != 0) {
      res = res - 1;
    }
    return res;
  }

  // NaN defination in IEEE 754-1985 is :
  //   - sign = either 0 or 1.
  //   - biased exponent = all 1 bits.
  //   - fraction = anything except all 0 bits (since all 0 bits represents infinity).
  // https://en.wikipedia.org/wiki/IEEE_754-1985#Representation_of_non-numbers
  fn isnan(val: f32) -> bool {
    let floatToUint: u32 = bitcast<u32>(val);
    return (floatToUint & 0x7fffffffu) > 0x7f800000u;
  }
  fn isnanVec4(val : vec4<f32>) -> vec4<bool> {
    return vec4<bool>(isnan(val[0]), isnan(val[1]), isnan(val[2]), isnan(val[3]));
  }
`;
function zne(e) {
  let t = "";
  switch (e) {
    case 0:
    case 1:
      t += `
        fn getOutputIndexFromCoords(coords : i32) -> i32 {
          return coords;
        }
        `;
      break;
    case 2:
      t += `
        fn getOutputIndexFromCoords(coords : vec2<i32>) -> i32 {
          return dot(coords, vec2<i32>(uniforms.outShapeStrides, 1));
        }
        `;
      break;
    case 3:
      t += `
        fn getOutputIndexFromCoords(coords : vec3<i32>) -> i32 {
          return dot(coords, vec3<i32>(uniforms.outShapeStrides.x, uniforms.outShapeStrides.y, 1));
        }
        `;
      break;
    case 4:
      t += `
        fn getOutputIndexFromCoords(coords : vec4<i32>) -> i32 {
          return dot(coords, vec4<i32>(
            uniforms.outShapeStrides.x, uniforms.outShapeStrides.y, uniforms.outShapeStrides.z, 1));
        }
        `;
      break;
    case 5:
      t += `
        fn getOutputIndexFromCoords(coords : vec5) -> i32 {
          return coords.x * uniforms.outShapeStrides.x +
              coords.y * uniforms.outShapeStrides.y +
              coords.z * uniforms.outShapeStrides.z +
              coords.w * uniforms.outShapeStrides.w +
              coords.u;
        }
        `;
      break;
    case 6:
      t += `
        fn getOutputIndexFromCoords(coords : vec6) -> i32 {
          return coords.x * uniforms.outShapeStrides.x +
              coords.y * uniforms.outShapeStrides.y +
              coords.z * uniforms.outShapeStrides.z +
              coords.w * uniforms.outShapeStrides.w +
              coords.u * uniforms.outShapeStrides.u +
              coords.v;
        }
        `;
      break;
    default:
      w.assert(false, () => `Unsupported ${e}D shape`);
      break;
  }
  return t;
}
function Mne(e, t, n) {
  let s = e.length, r = hd(t, n), a;
  if (n ? a = `fn setOutputAtIndex(flatIndex : i32, value : vec4<f32>) {
      result[flatIndex] = ${r}(value);
    }
    fn setOutputAtIndexI32(flatIndex : i32, value : vec4<i32>) {
      result[flatIndex] = ${r}(value);
    }` : a = `fn setOutputAtIndex(flatIndex : i32, value : f32) {
      result[flatIndex] = ${r}(value);
    }
    fn setOutputAtIndexI32(flatIndex : i32, value : i32) {
      result[flatIndex] = ${r}(value);
    }`, s >= 2) {
    let i = ["d0", "d1", "d2", "d3", "d4", "d5"].slice(0, s), o = Ut(s);
    n ? a += `
      fn setOutputAtCoords(${i.map((u) => `${u} : i32`).join(", ")}, value : vec4<f32>) {
        let flatIndex = getOutputIndexFromCoords(${o}(${i.join(", ")}));
        setOutputAtIndex(flatIndex / 4, value);
      }
      fn setOutputAtCoordsI32(${i.map((u) => `${u} : i32`).join(", ")}, value : vec4<i32>) {
        let flatIndex = getOutputIndexFromCoords(${o}(${i.join(", ")}));
        setOutputAtIndexI32(flatIndex / 4, value);
      }
    ` : a += `
      fn setOutputAtCoords(${i.map((u) => `${u} : i32`).join(", ")}, value : f32) {
        let flatIndex = getOutputIndexFromCoords(${o}(${i.join(", ")}));
        setOutputAtIndex(flatIndex, value);
      }
      fn setOutputAtCoordsI32(${i.map((u) => `${u} : i32`).join(", ")}, value : i32) {
        let flatIndex = getOutputIndexFromCoords(${o}(${i.join(", ")}));
        setOutputAtIndexI32(flatIndex, value);
      }
    `;
  }
  return a;
}
function Lne(e, t, n, s) {
  let r = Bne(e, n);
  return e.shape.length <= t.length && (r += Vne(e, t, n, s)), r;
}
function Bne(e, t) {
  let n = e.name, s = e.shape.length, r = Ut(s), a = "get" + n.charAt(0).toUpperCase() + n.slice(1), i = ["d0", "d1", "d2", "d3", "d4", "d5"].slice(0, s), o = i.map((c) => `${c} : i32`).join(", ");
  if (s < 1)
    return t ? `
        fn ${a}() -> vec4<f32> {
          return vec4<f32>(${n}[0]);
        }
      ` : `
      fn ${a}() ->f32 {
        return f32(${n}[0]);
      }
    `;
  let u = `uniforms.${n.charAt(0).toLowerCase() + n.slice(1)}Shape`, l = `${s}D`;
  return s === 0 && (l = "1D"), t ? `
      fn ${a}(${o}) -> vec4<f32> {
        return vec4<f32>(${n}[getIndexFromCoords${l}(${r}(${i.join(",")}),
          ${u}) / 4]);
      }
      ` : `
    fn ${a}(${o}) -> f32 {
      return f32(${n}[getIndexFromCoords${l}(${r}(${i.join(",")}),
        ${u})]);
    }
   `;
}
function Vne(e, t, n, s) {
  let r = e.name, a = r.charAt(0).toUpperCase() + r.slice(1), i = "get" + a + "ByOutput", o = e.shape.length, u = t.length, l = Ut(u);
  if (w.arraysEqual(e.shape, t) && s)
    return n ? `
        fn ${i}Index(globalIndex : i32) -> vec4<f32> {
          return vec4<f32>(${r}[globalIndex]);
        }

        fn ${i}Coords(coords : ${l}) -> vec4<f32> {
          return vec4<f32>(${r}[${u > 1 ? "getOutputIndexFromCoords(coords)" : "coords"} / 4]);
        }
        ` : `
      fn ${i}Index(globalIndex : i32) -> f32 {
        return f32(${r}[globalIndex]);
      }

      fn ${i}Coords(coords : ${l}) -> f32 {
        return f32(${r}[${u > 1 ? "getOutputIndexFromCoords(coords)" : "coords"}]);
      }
      `;
  let c = C.getBroadcastDims(e.shape, t), p = u - o, d = "";
  if (o === 0)
    return n ? `
      fn ${i}Index(globalIndex : i32) -> vec4<f32> {
        return get${a}();
      }

      fn ${i}Coords(coords : ${l}) -> vec4<f32> {
        return get${a}();
      }
    ` : `
      fn ${i}Index(globalIndex : i32) -> f32{
        return get${a}();
      }

      fn ${i}Coords(coords : ${l}) -> f32{
        return get${a}();
      }
    `;
  u < 2 && c.length >= 1 ? d = "coords = 0;" : d = c.map((g) => `coords.${pr(g + p)} = 0;`).join(`
`);
  let h = "";
  if (u < 2 && o > 0)
    h = "coords";
  else if (u > 1) {
    let g = Ut(o), b = e.shape.map((y, v) => `coords.${pr(v + p)}`).join(", ");
    h = `${g}(${b})`;
  } else
    h = "coords";
  let f = `uniforms.${r.charAt(0).toLowerCase() + r.slice(1)}Shape`, m = `${o}D`;
  return n ? `
      fn ${i}Index(globalIndex : i32) -> vec4<f32> {
        var coords = getCoordsFromIndex(globalIndex);
        ${d}
        return ${r}[getIndexFromCoords${m}(${h}, ${f}) / 4];
      }

      fn ${i}Coords(coordsIn : ${l}) -> vec4<f32> {
        var coords = coordsIn;
        ${d}
        return ${r}[getIndexFromCoords${m}(${h}, ${f}) / 4];
      }
    ` : `
    fn ${i}Index(globalIndex : i32) -> f32 {
      var coords = getCoordsFromIndex(globalIndex);
      ${d}
      return f32(${r}[getIndexFromCoords${m}(${h}, ${f})]);
    }

    fn ${i}Coords(coordsIn : ${l}) -> f32 {
      var coords = coordsIn;
      ${d}
      return f32(${r}[getIndexFromCoords${m}(${h}, ${f})]);
    }
  `;
}
function Wne(e, t) {
  let { x: n, y: s = [], z: r = [] } = t, a = e.length;
  if (n.length === a)
    return [`fn getOutputCoords() -> ${Ut(a)}{
      let globalIndex = getGlobalIndex();
      return getCoordsFromIndex(globalIndex);
    }
    `, a];
  let i = "", o = [n, s, r], u = 0;
  for (let d = 0; d < o.length; d++) {
    let h = o[d];
    if (h.length !== 0)
      if (u += h.length, h.length === 1)
        i += `let d${h[0]} = i32(globalId[${d}]);`;
      else {
        let f = One(h, "uniforms.outShape");
        i += `var index${d} = i32(globalId[${d}]);`;
        for (let m = 0; m < f.length; m++)
          i += `let d${h[m]} = index${d} / ${f[m]};`, m === f.length - 1 ? i += `let d${h[m + 1]} = index${d} - d${h[m]} * ${f[m]};` : i += `index${d} = index${d} - d${h[m]} * ${f[m]};`;
      }
  }
  let l = [];
  for (let d = 0; d < u; d++)
    l.push(`d${d}`);
  let c = Ut(u), p = `fn getOutputCoords() -> ${c} {
    ${i}
  `;
  return l.length === 0 ? p += `return ${c}(0); }` : p += `return ${c}(${l.join(",")}); }`, [p, u];
}
function Ow(e) {
  let t = e.length;
  if (t <= 1)
    return "fn getCoordsFromIndex(index : i32) -> i32 { return index; }";
  let n = w.computeStrides(e), s = Ut(t), r = [];
  for (let i = 0; i < t; i++)
    r.push(`d${i}`);
  if (n.length === 1)
    return `    fn getCoordsFromIndex(index : i32) -> vec2<i32> {
      let d0 = index / uniforms.outShapeStrides; let d1 = index - d0 * uniforms.outShapeStrides;
      return vec2<i32>(d0, d1);
    }`;
  let a;
  return a = "var index2 = index;" + n.map((i, o) => {
    let u = `let ${r[o]} = index2 / uniforms.outShapeStrides.${pr(o)}`, l = o === n.length - 1 ? `let ${r[o + 1]} = index2 - ${r[o]} * uniforms.outShapeStrides.${pr(o)}` : `index2 = index2 - ${r[o]} * uniforms.outShapeStrides.${pr(o)}`;
    return `${u}; ${l};`;
  }).join(""), `
    fn getCoordsFromIndex(index : i32) -> ${s} {
      ${a}
      return ${s}(${r.join(",")});
    }
  `;
}
var N2 = {};
Ee(N2, { ArrayBufferToTypedArray: () => $2, GPUBytesPerElement: () => fd, computeDispatch: () => _e, computeWorkGroupSizeForConv2d: () => Rv, computeWorkGroupSizeForMatMul: () => T2, computeWorkPerThreadForConv2d: () => Dv, flatDispatchLayout: () => Be, isWebGPUSupported: () => Fv, tilesFitEvenlyIntoShape: () => js });
var sa = (e) => {
  let t = 1;
  for (let n = 0; n < e.length; n++)
    t *= e[n];
  return t;
};
function js(e, t) {
  if (e.length !== t.length)
    throw new Error(`Cannot compute whether rank ${e.length} tiles fit evenly into rank ${t.length} shape - ranks must match.`);
  return t.every((n, s) => n % e[s] === 0);
}
function _e(e, t, n = [1, 1, 1], s = [1, 1, 1]) {
  let [r, a, i] = [Math.ceil(sa(e.x.map((o) => t[o])) / (n[0] * s[0])), e.y ? Math.ceil(sa(e.y.map((o) => t[o])) / (n[1] * s[1])) : 1, e.z ? Math.ceil(sa(e.z.map((o) => t[o])) / (n[2] * s[2])) : 1];
  return [r, a, i];
}
function Rv(e, t) {
  let n = sa(e.x.map((r) => t[r])), s = sa(e.y.map((r) => t[r]));
  return n <= 4 ? [4, 16, 1] : s <= 4 ? [16, 4, 1] : [16, 16, 1];
}
function T2(e, t, n) {
  return e === 1 ? [32, 1, 1] : n === 1 ? [1, 32, 1] : [8, 8, 1];
}
function Dv(e, t) {
  let n = sa(e.x.map((r) => t[r])), s = sa(e.y.map((r) => t[r]));
  return n <= 4 ? [1, 2, 1] : s <= 4 ? [2, 1, 1] : [2, 2, 1];
}
function Be(e) {
  return { x: e.map((t, n) => n) };
}
function fd(e) {
  if (e === "float32" || e === "int32" || e === "bool" || e === "string")
    return 4;
  if (e === "complex64")
    return 8;
  throw new Error(`Unknown dtype ${e}`);
}
function $2(e, t) {
  if (t === "float32")
    return new Float32Array(e);
  if (t === "int32")
    return new Int32Array(e);
  if (t === "bool" || t === "string")
    return Uint8Array.from(new Int32Array(e));
  throw new Error(`Unknown dtype ${t}`);
}
function Fv() {
  return (typeof window != "undefined" || typeof WorkerGlobalScope != "undefined") && !!navigator.gpu;
}
function _2(e, t, n, s) {
  return w.assert(s % 4 === 0 && e[0] === 4, () => "tileInner must be divisible by 4. And ColPerThread must be 4"), `
  var<workgroup> mm_Asub : array<array<vec4<f32>, ${s / e[0]}>, ${t}>;
  var<workgroup> mm_Bsub : array<array<vec4<f32>, ${n / e[0]}>, ${s}>;

  let RowPerThread = ${e[1]};
  let ColPerThread = ${e[0]};
  let TileInner = ${s};

  ${Ii()}

    let tileRow = ${t === 1 ? "0" : "i32(localId.y) * RowPerThread"};
    let tileCol = i32(localId.x);

    let globalRow = ${t === 1 ? "0" : "i32(globalId.y) * RowPerThread"};
    let globalCol = i32(globalId.x);
    let numTiles = (uniforms.dimInner - 1) / TileInner + 1;

    var acc: array<vec4<f32>, RowPerThread>;
    var ACached : vec4<f32>;
    var BCached : array<vec4<f32>, 4>;

    // Loop over shared dimension.
    var globalColA = tileCol;
    let RowPerThreadB = TileInner / i32(workGroupSizeY);
    let tileRowB = i32(localId.y) * RowPerThreadB;
    for (var t = 0; t < numTiles; t = t + 1) {
        // Load one tile of A into local memory.
        for (var innerRow = 0; innerRow < RowPerThread; innerRow = innerRow + 1) {
            let inputRow = tileRow + innerRow;
            let inputCol = tileCol;
            mm_Asub[inputRow][inputCol] = mm_readA(globalRow + innerRow, globalColA, globalId);
        }
        globalColA = globalColA + TileInner / ColPerThread;

        // Load one tile of B into local memory.
        for (var innerRow = 0; innerRow < RowPerThreadB; innerRow = innerRow + 1) {
            let inputRow = tileRowB + innerRow;
            let inputCol = tileCol;
            mm_Bsub[inputRow][inputCol] = mm_readB(t * TileInner + inputRow, globalCol, globalId);
        }

        workgroupBarrier();

        // Compute acc values for a single thread.
        for (var k = 0; k < TileInner / ColPerThread; k = k + 1) {
            BCached[0] = mm_Bsub[k * ColPerThread][tileCol];
            BCached[1] = mm_Bsub[k * ColPerThread + 1][tileCol];
            BCached[2] = mm_Bsub[k * ColPerThread + 2][tileCol];
            BCached[3] = mm_Bsub[k * ColPerThread + 3][tileCol];

            for (var i = 0; i < RowPerThread; i = i + 1) {
                ACached = mm_Asub[tileRow + i][k];
                acc[i] = BCached[0] * ACached.x + acc[i];
                acc[i] = BCached[1] * ACached.y + acc[i];
                acc[i] = BCached[2] * ACached.z + acc[i];
                acc[i] = BCached[3] * ACached.w + acc[i];
            }
        }

        workgroupBarrier();
    }

    for (var innerRow = 0; innerRow < RowPerThread; innerRow = innerRow + 1) {
        mm_write(globalRow + innerRow,
                 globalCol,
                 acc[innerRow], globalId);
    }
  }`;
}
var Une = class {
  constructor(e, t, n, s, r, a = null, i = null, o = null) {
    this.variableNames = ["A", "B"], this.uniforms = "dimAOuter : i32, dimBOuter : i32, dimInner : i32,", this.workGroupSize = [8, 8, 1], this.isVec4 = true, this.outputShape = t, this.dispatchLayout = { x: [2], y: [1], z: [0] }, t[1] === 1 ? this.elementsPerThread = [4, 1, 1] : this.elementsPerThread = [4, 4, 1], this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, this.elementsPerThread);
    let u = a != null, l = o != null;
    u && this.variableNames.push("bias"), l && this.variableNames.push("preluActivationWeights"), this.tileAOuter = t[1] === 1 ? 1 : this.workGroupSize[1] * this.elementsPerThread[1], this.tileBOuter = this.workGroupSize[0] * this.elementsPerThread[0], this.tileInner = this.tileBOuter, this.aShape = e, this.addBias = u, this.activation = i, this.hasPreluActivationWeights = l, this.batchAEqualOne = s, this.batchBEqualOne = r, [this.fitA, this.fitB] = this.getShapeFit(), this.shaderKey = `matMulPackedVec4_${this.activation}_${this.fitA}_${this.fitB}_${this.elementsPerThread}_${this.batchAEqualOne}_${this.batchBEqualOne}`;
  }
  getShapeFit() {
    let e = this.aShape[2], t = this.outputShape[2], n = [this.outputShape[0], e, t], s = [this.tileAOuter, this.tileInner], r = [this.tileInner, this.tileBOuter];
    return [js(s, this.aShape.slice(1)), js(r, n.slice(1))];
  }
  getUserCode() {
    let e = this.fitA ? "return A[batch * batchASize + row * uniforms.dimInner / 4 + col]" : `if (coordsInBounds2D(vec2<i32>(row, col * 4), vec2<i32>(uniforms.dimAOuter, uniforms.dimInner))) {
            return A[batch * batchASize + row * uniforms.dimInner / 4 + col];
        }
        return vec4<f32>(0.0)`, t = this.fitB ? "return B[batch * batchBSize + row * uniforms.dimBOuter / 4 + col]" : `if(coordsInBounds2D(vec2<i32>(row, col * 4), vec2<i32>(uniforms.dimInner, uniforms.dimBOuter))) {
             return B[batch * batchBSize + row * uniforms.dimBOuter / 4 + col];
        }
        return vec4<f32>(0.0)`, n = "", s = "";
    if (this.activation) {
      let i = Fr(this.activation, this.isVec4);
      this.hasPreluActivationWeights ? n = `fn activation(a : vec4<f32>, outCoord : vec3<i32>) -> vec4<f32> {
                  let b = getPreluActivationWeightsByOutputCoords(outCoord);
                  ${i}
                }` : n = `
            fn activation(a : vec4<f32>, outCoord : vec3<i32>) -> vec4<f32> {
              ${i}
            }`, s = "value = activation(value, outCoord);";
    }
    let r = this.addBias ? "value = value + getBiasByOutputCoords(outCoord);" : "";
    return `
      ${n}
      fn mm_readA(row : i32, col : i32,  globalId : vec3<u32>) -> vec4<f32> {
        ${this.batchAEqualOne ? `
          let batchASize = 0;
          let batch = 0;
        ` : `
          let batchASize = uniforms.aShape[1] * uniforms.aShape[2] / 4;
          let batch = i32(globalId.z);
        `}

        ${e};
      }

      fn mm_readB(row : i32, col : i32,  globalId : vec3<u32>) -> vec4<f32> {
        ${this.batchBEqualOne ? `
          let batchBSize = 0;
          let batch = 0;
          ` : `
          let batchBSize = uniforms.bShape[1] * uniforms.bShape[2] / 4;
          let batch = i32(globalId.z);
       `}
        ${t};
      }

      fn mm_write(row : i32, col : i32, valueIn : vec4<f32>, globalId : vec3<u32>) {
        if (row < uniforms.aShape[1] && col * 4 < uniforms.bShape[2])
        {
          var value = valueIn;
          let batch = i32(globalId.z);
          let outCoord = vec3<i32>(batch, row, col * 4);
          ${r}
          ${s}
          setOutputAtCoords(outCoord[0], outCoord[1], outCoord[2], value);
        }
      }
      ${_2(this.elementsPerThread, this.tileAOuter, this.tileBOuter, this.tileInner)}
    `;
  }
};
function Ov(e, t) {
  let n = t[1] * e[1], s = t[0] * e[0], r = n > s ? n : s;
  return `
    var<workgroup> mm_Asub : array<array<f32, ${r}>, ${n}>;
    var<workgroup> mm_Bsub : array<array<f32, ${s}>, ${r}>;
    ${Ii()}
      let tileRow = i32(localId.y) * ${e[1]};
      let tileCol = i32(localId.x) * ${e[0]};

      let globalRow = i32(globalId.y) * ${e[1]};
      let globalCol = i32(globalId.x) * ${e[0]};

      let numTiles = (uniforms.dimInner - 1) / ${r} + 1;

      var acc : array<array<f32, ${e[0]}>, ${e[1]}>;
      var ACached : f32;
      var BCached : array<f32, ${e[0]}>;

      // Without this initialization strange values show up in acc.
      for (var innerRow = 0; innerRow < ${e[1]}; innerRow = innerRow + 1) {
        for (var innerCol = 0; innerCol < ${e[0]}; innerCol = innerCol + 1) {
          acc[innerRow][innerCol] = 0.0;
        }
      }

      let ColPerThreadA = ${r} / ${t[0]};
      let tileColA = i32(localId.x) * ColPerThreadA;
      let RowPerThreadB = ${r} / ${t[1]};
      let tileRowB = i32(localId.y) * RowPerThreadB;

      // Loop over shared dimension.
      for (var t = 0; t < numTiles; t = t + 1) {
        // Load one tile of A into local memory.
        for (var innerRow = 0; innerRow < ${e[1]}; innerRow = innerRow + 1) {
          for (var innerCol = 0; innerCol < ColPerThreadA; innerCol = innerCol + 1) {
            let inputRow = tileRow + innerRow;
            let inputCol = tileColA + innerCol;

            mm_Asub[inputRow][inputCol] = mm_readA(
                globalRow + innerRow,
                t * ${r} + inputCol, globalId);
          }
        }
        // Load one tile of B into local memory.
        for (var innerRow = 0; innerRow < RowPerThreadB; innerRow = innerRow + 1) {
          for (var innerCol = 0; innerCol < ${e[0]}; innerCol = innerCol + 1) {
            let inputRow = tileRowB + innerRow;
            let inputCol = tileCol + innerCol;

            mm_Bsub[inputRow][inputCol] = mm_readB(
              t * ${r} + inputRow,
              globalCol + innerCol, globalId);
          }
        }

        workgroupBarrier();

        // Compute acc values for a single thread.
        for (var k = 0; k < ${r}; k = k + 1) {
          for (var inner = 0; inner < ${e[0]}; inner = inner + 1) {
            BCached[inner] = mm_Bsub[k][tileCol + inner];
          }

          for (var innerRow = 0; innerRow < ${e[1]}; innerRow = innerRow + 1) {
            ACached = mm_Asub[tileRow + innerRow][k];
            for (var innerCol = 0; innerCol < ${e[0]}; innerCol = innerCol + 1) {
              acc[innerRow][innerCol] = acc[innerRow][innerCol] + ACached * BCached[innerCol];
            }
          }
        }

        workgroupBarrier();
      }

      for (var innerRow = 0; innerRow < ${e[1]}; innerRow = innerRow + 1) {
        for (var innerCol = 0; innerCol < ${e[0]}; innerCol = innerCol + 1) {

          if ((globalCol + innerCol) < uniforms.dimBOuter &&
              (globalRow + innerRow) < uniforms.dimAOuter) {
            mm_write(globalRow + innerRow,
                     globalCol + innerCol,
                     acc[innerRow][innerCol], globalId);
          }
        }
      }
    }
  `;
}
function Gne(e) {
  return `
    let TileSize = ${e[0] * 4};
    var<workgroup> mm_Asub : array<vec4<f32>, ${e[0]}>;

    ${Ii()}
      let tileCol = i32(localId.x);
      let globalCol = i32(globalId.x);
      let globalRow = i32(globalId.y);

      let numTiles = (uniforms.dimInner - 1) / TileSize + 1;

      // Without this initialization strange values show up in acc.
      var acc = 0.0;

      // Loop over shared dimension.
      for (var t = 0; t < numTiles; t = t + 1) {
        // Load one tile of A into local memory.
        let colA = t * TileSize + tileCol * 4;
        mm_Asub[tileCol] = vec4<f32>(mm_readA(globalRow, colA, globalId),
                                mm_readA(globalRow, colA + 1, globalId),
                                mm_readA(globalRow, colA + 2, globalId),
                                mm_readA(globalRow, colA + 3, globalId));
        workgroupBarrier();

        // Compute acc values for a single thread.
        for (var k = 0; k < TileSize / 4; k = k + 1) {
          let rowB = t * TileSize + k * 4;
          let BCached = vec4<f32>(mm_readB(rowB, globalCol, globalId),
                              mm_readB(rowB + 1, globalCol, globalId),
                              mm_readB(rowB + 2, globalCol, globalId),
                              mm_readB(rowB + 3, globalCol, globalId));

          let ACached = mm_Asub[k];
          acc = acc + dot(ACached, BCached);
        }

        workgroupBarrier();
      }

      if (globalRow < uniforms.dimAOuter && globalCol < uniforms.dimBOuter) {
        mm_write(globalRow, globalCol, acc, globalId);
      }
    }
  `;
}
var Hne = class {
  constructor(e, t, n, s, r, a = false, i = false, o = null, u = null, l = null) {
    this.variableNames = ["A", "B"], this.uniforms = "dimAOuter : i32, dimBOuter : i32, dimInner : i32,", this.workGroupSize = [16, 16, 1], this.outputShape = t, this.dispatchLayout = { x: [2], y: [1], z: [0] };
    let c = a ? e[1] : e[2];
    this.workGroupSize = T2(t[1], c, t[2]), (t[1] === 1 || t[2] === 1) && (n = 1), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [n, n, 1]), w.arraysEqual(this.dispatch, [1, 1, 1]) && (n = 1, this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [n, n, 1]));
    let p = o != null, d = l != null;
    p && this.variableNames.push("bias"), d && this.variableNames.push("preluActivationWeights"), this.workPerThread = n, this.aShape = e, this.transposeA = a, this.transposeB = i, this.addBias = p, this.activation = u, this.hasPreluActivationWeights = d, this.batchAEqualOne = s, this.batchBEqualOne = r;
    let h = this.outputShape[2], f = this.transposeB ? [this.outputShape[0], h, c] : [this.outputShape[0], c, h];
    [this.fitA, this.fitB] = this.getShapeFit(f), this.shaderKey = `matMulPacked_${this.workPerThread}_${a}_${i}_${this.activation}_${this.fitA}_${this.fitB}_${this.outputShape[1] > 1}_${this.batchAEqualOne}_${this.batchBEqualOne}`;
  }
  getShapeFit(e) {
    let t = this.workGroupSize[1] * this.workPerThread, n = this.workGroupSize[0] * this.workPerThread, s = t > n ? t : n;
    this.outputShape[1] === 1 && (s *= 4), w.assert(s % this.workGroupSize[0] === 0 && s % this.workGroupSize[1] === 0, () => "tileInner must be multiple of workgroupsize.x and workgroupsize.y");
    let r = [t, s], a = [s, n];
    return [js(r, this.aShape.slice(1)), js(a, e.slice(1))];
  }
  getUserCode() {
    let e;
    this.transposeA === false ? e = this.fitA ? "return A[batch * batchASize + row * uniforms.dimInner + col];" : `if(coordsInBounds2D(vec2<i32>(row, col), vec2<i32>(uniforms.dimAOuter, uniforms.dimInner))) {
             return A[batch * batchASize + row * uniforms.dimInner + col];
           }
           return 0.0;` : e = this.fitA ? "return A[batch * batchASize + col * uniforms.dimAOuter + row];" : `if(coordsInBounds2D(vec2<i32>(row, col), vec2<i32>(uniforms.dimAOuter, uniforms.dimInner))) {
             return A[batch* batchASize + col * uniforms.dimAOuter + row];
           }
           return 0.0;`;
    let t;
    this.transposeB === false ? t = this.fitB ? "return B[batch * batchBSize + row * uniforms.dimBOuter + col];" : `if(coordsInBounds2D(vec2<i32>(row, col), vec2<i32>(uniforms.dimInner, uniforms.dimBOuter))) {
             return B[batch * batchBSize + row * uniforms.dimBOuter + col];
           }
           return 0.0;` : t = this.fitB ? "return B[batch * batchBSize + col * uniforms.dimInner + row];" : `if(coordsInBounds2D(vec2<i32>(row, col), vec2<i32>(uniforms.dimInner, uniforms.dimBOuter))) {
             return B[batch * batchBSize + col * uniforms.dimInner + row];
           }
           return 0.0;`;
    let n = "", s = "";
    if (this.activation) {
      let i = Fr(this.activation, false);
      this.hasPreluActivationWeights ? n = `fn activation(a : f32, outCoord : vec3<i32>) -> f32 {
               let b = getPreluActivationWeightsByOutputCoords(outCoord);
               ${i}
            }` : n = `
              fn activation(a : f32, outCoord : vec3<i32>) -> f32 {
                ${i}
              }
            `, s = "value = activation(value, outCoord);";
    }
    let r = this.addBias ? "value = value + getBiasByOutputCoords(outCoord);" : "";
    return `
      ${n}

      fn mm_readA(row : i32, col : i32,  globalId : vec3<u32>) -> f32 {
        ${this.batchAEqualOne ? `
        let batch = 0;
        let batchASize = 0;
        ` : `
        let batch = i32(globalId.z);
        let batchASize = uniforms.aShape[1] * uniforms.aShape[2];
        `}
        ${e}
      }

      fn mm_readB(row : i32, col : i32,  globalId : vec3<u32>) -> f32 {
        ${this.batchBEqualOne ? `
        let batch = 0;
        let batchBSize = 0;
        ` : `
        let batch = i32(globalId.z);
        let batchBSize = uniforms.bShape[1] * uniforms.bShape[2];
        `}
        ${t}
      }

      fn mm_write(row : i32, col : i32, valueIn : f32, globalId : vec3<u32>) {
        var value = valueIn;
        let batch = i32(globalId.z);
        let outCoord = vec3<i32>(batch, row, col);
        ${r}
        ${s}
        setOutputAtCoords(batch, row, col, value);
      }
      ${this.outputShape[1] > 1 ? Ov([this.workPerThread, this.workPerThread, 1], this.workGroupSize) : Gne(this.workGroupSize)}
    `;
  }
};
function qne() {
  return `
    var<workgroup> sumValues : array<f32, workGroupSizeX>;
    ${Ii()}
      let coords = getOutputCoords();
      let batch = coords[0];
      let row = coords[1];
      let col = coords[2];
      var sum = 0.0;
      let Length = uniforms.dimInner;
      for (var k = i32(localId.x); k < Length; k = k + i32(workGroupSizeX)) {
        let dataA = mm_readA(batch, row, k);
        let dataB = mm_readB(batch, k, col);
        sum = sum + dataA * dataB;
      }
      sumValues[localId.x] = sum;
      workgroupBarrier();

      for(var currentSize = workGroupSizeX / 2u; currentSize > 1u;
          currentSize = currentSize / 2u) {
        if (localId.x < currentSize)
        {
          sumValues[localId.x] = sumValues[localId.x] + sumValues[localId.x + currentSize];
        }
        workgroupBarrier();
      }

      if (localId.x == 0u) {
        sum = sumValues[0] + sumValues[1];
        mm_write(batch, row, col, sum);
      }
    }
  `;
}
var jne = class {
  constructor(e, t, n, s = false, r = false, a = null, i = null, o = null) {
    this.variableNames = ["A", "B"], this.uniforms = "dimAOuter : i32, dimBOuter : i32, dimInner : i32,", this.workGroupSize = [256, 1, 1], this.outputShape = e, this.dispatchLayout = { x: [], y: [1, 2], z: [0] }, this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize);
    let u = a != null, l = o != null;
    u && this.variableNames.push("bias"), l && this.variableNames.push("preluActivationWeights"), this.transposeA = s, this.transposeB = r, this.addBias = u, this.activation = i, this.hasPreluActivationWeights = l, this.batchAEqualOne = t, this.batchBEqualOne = n, this.shaderKey = `matMulReduce_${this.activation}_${s}_${r}_${this.batchAEqualOne}_${this.batchBEqualOne}`;
  }
  getUserCode() {
    let e;
    this.transposeA === false ? e = "return f32(A[batch * batchASize + row * uniforms.dimInner + col]);" : e = "return f32(A[batch * batchASize + col * uniforms.dimAOuter + row]);";
    let t;
    this.transposeB === false ? t = "return f32(B[batch * batchBSize + row * uniforms.dimBOuter + col]);" : t = "return f32(B[batch * batchBSize + col * uniforms.dimInner + row]);";
    let n = "", s = "";
    if (this.activation) {
      let i = Fr(this.activation, false);
      this.hasPreluActivationWeights ? n = `fn activation(a : f32, outCoord : vec3<i32>) -> f32 {
               let b = getPreluActivationWeightsByOutputCoords(outCoord);
               ${i}
            }` : n = `
              fn activation(a : f32, outCoord : vec3<i32>) -> f32 {
                ${i}
              }
            `, s = "value = activation(value, outCoord);";
    }
    let r = this.addBias ? "value = value + getBiasByOutputCoords(outCoord);" : "";
    return `
      ${n}

      fn mm_readA(batchIn: i32, row : i32, col : i32) -> f32 {
        ${this.batchAEqualOne ? `
          let batchASize = 0;
          let batch = 0;
          ` : `
          let batchASize = uniforms.aShape[1] * uniforms.aShape[2];
          let batch = batchIn;
          `}
        ${e}
      }

      fn mm_readB(batchIn: i32, row : i32, col : i32) -> f32 {
        ${this.batchBEqualOne ? `
          let batch = 0;
          let batchBSize = 0;
          ` : `
          let batch = batchIn;
          let batchBSize = uniforms.bShape[1] * uniforms.bShape[2];
          `}
        ${t}
      }

      fn mm_write(batch: i32, row : i32, col : i32, valueIn : f32) {
        var value = valueIn;
        let outCoord = vec3<i32>(batch, row, col);
        ${r}
        ${s}
        setOutputAtCoords(batch, row, col, value);
      }
      ${qne()}
    `;
  }
};
function Kne(e) {
  let t = e[1] / 2, n = e[0], s = t > n ? t : n;
  return `
  var<workgroup> mm_Asub1 : array<array<f32, ${s}>, ${t}>;
  var<workgroup> mm_Bsub1 : array<array<f32, ${n}>, ${s}>;
  var<workgroup> mm_Asub2 : array<array<f32, ${s}>, ${t}>;
  var<workgroup> mm_Bsub2 : array<array<f32, ${n}>, ${s}>;

  // If the output size is small for matrix multiplication, avoid to use vec4
  // and handle some elements per thread to optimally utilize the ALU.
  // Introduces two shared memory buffers, some logical threads could handle
  // arithmetic operations and others handle IO operations between barrier api,
  // makes ALUs and load/store units work simultaneously, could improves
  // the performance.
  ${Ii()}
    let tileRow = i32(localId.y);
    let tileCol = i32(localId.x);
    let globalRow = i32(globalId.y);
    let globalCol = i32(globalId.x);

    // uniforms.dimInner should be greater than 0.
    let numTiles = (uniforms.dimInner - 1) / ${s} + 1;
    var acc = 0.0;

    var globalColA = tileCol;
    var globalRowB = tileRow;
    for (var t = 0; t < numTiles; t = t + 1) {
      if (t == 0) {
        if (tileRow < ${t}) {
          // Load one tile of A and B into local memory.
          // globalRow is always greater than or equal tileRow.
          mm_Asub1[tileRow][tileCol] =
              mm_readA((globalRow - tileRow) / 2 + tileRow, globalColA, globalId);
          globalColA = globalColA + ${s};
          mm_Bsub1[tileRow][tileCol] = mm_readB(globalRowB, globalCol, globalId);
          globalRowB = globalRowB + ${s};
        }
      } else {
        if (tileRow < ${t}) {
          // Load one tile of A and B into local memory.
          // globalRow is always greater than or equal tileRow.
          mm_Asub1[tileRow][tileCol] =
              mm_readA((globalRow - tileRow) / 2 + tileRow, globalColA, globalId);
          globalColA = globalColA + ${s};
          mm_Bsub1[tileRow][tileCol] = mm_readB(globalRowB, globalCol, globalId);
          globalRowB = globalRowB + ${s};
        } else {
          // Compute acc values for a single thread.
          for (var k = 0; k < ${s}; k = k + 1) {
            let subRow = tileRow - ${t};
            if (subRow < 0) {
              continue;
            }
            acc = acc + mm_Asub2[subRow][k] * mm_Bsub2[k][tileCol];
          }
        }
      }
      workgroupBarrier();
      if (t != 0) {
        t = t + 1;
      }

      if (t < numTiles) {
        if (tileRow < ${t}) {
          // Load one tile of A and B into local memory.
          // globalRow is always greater than or equal tileRow.
          mm_Asub2[tileRow][tileCol] =
              mm_readA((globalRow - tileRow) / 2 + tileRow, globalColA, globalId);
          globalColA = globalColA + ${s};
          mm_Bsub2[tileRow][tileCol] = mm_readB(globalRowB, globalCol, globalId);
          globalRowB = globalRowB + ${s};
        } else {
          // Compute acc values for a single thread.
          for (var k = 0; k < ${s}; k = k + 1) {
            let subRow = tileRow - ${t};
            if (subRow < 0) {
              continue;
            }
            acc = acc + mm_Asub1[subRow][k] * mm_Bsub1[k][tileCol];
          }
        }
      }
      workgroupBarrier();
    }
    let writeCol = (globalRow - tileRow) / 2 + tileRow - ${t};
    if (tileRow >= ${t} && writeCol >= 0) {
      mm_write(writeCol, globalCol, acc, globalId);
    }
  }
  `;
}
var Xne = class {
  constructor(e, t, n, s = null, r = null, a = null) {
    this.variableNames = ["A", "B"], this.uniforms = "dimAOuter : i32, dimBOuter : i32, dimInner : i32,", this.workGroupSize = [8, 16, 1], w.assert(e[1] <= 16 || t[2] <= 16, () => "This program can be only used when A width or B Height are small"), this.outputShape = n, this.dispatchLayout = { x: [2], y: [1], z: [0] }, this.dispatch = [Math.ceil(n[2] / this.workGroupSize[0]), Math.ceil(n[1] * 2 / this.workGroupSize[1]), n[0]];
    let i = s != null;
    i && this.variableNames.push("bias");
    let o = a != null;
    o && this.variableNames.push("preluActivationWeights"), this.addBias = i, this.activation = r, this.hasPreluActivationWeights = o, this.batchAEqualOne = e[0] === 1, this.batchBEqualOne = t[0] === 1, this.shaderKey = `matMulSmallOutputSize_${this.activation}_${this.batchAEqualOne}_${this.batchBEqualOne}`;
  }
  getUserCode() {
    let e = `if (coordsInBounds2D(vec2<i32>(row, col), vec2<i32>(uniforms.dimAOuter, uniforms.dimInner))) {
          return A[batch * batchASize + row * uniforms.dimInner + col];
        }
        return 0.0;`, t = `if (coordsInBounds2D(vec2<i32>(row, col), vec2<i32>(uniforms.dimInner, uniforms.dimBOuter))) {
           return B[batch * batchBSize + row * uniforms.dimBOuter + col];
         }
         return 0.0;`, n = "", s = "";
    if (this.activation) {
      let i = Fr(this.activation, false);
      this.hasPreluActivationWeights ? n = `fn activation(a : f32, outCoord : vec3<i32>) -> f32 {
            let b = getPreluActivationWeightsByOutputCoords(outCoord);
            ${i}
            }` : n = `fn activation(a : f32, outCoord : vec3<i32>) -> f32 {
            ${i}
        }`, s = "value = activation(value, outCoord);";
    }
    let r = this.addBias ? "value = value + getBiasByOutputCoords(outCoord);" : "";
    return `
      ${n}

      fn mm_readA(row : i32, col : i32,  globalId : vec3<u32>) -> f32 {
        ${this.batchAEqualOne ? `
          let batch = 0;
          let batchASize = 0;
          ` : `
          let batchASize = uniforms.aShape[1] * uniforms.aShape[2];
          let batch = i32(globalId.z);
          `}
        ${e}
      }
      fn mm_readB(row : i32, col : i32,  globalId : vec3<u32>) -> f32 {
        ${this.batchBEqualOne ? `
          let batch = 0;
          let batchBSize = 0;
          ` : `
          let batch = i32(globalId.z);
          let batchBSize = uniforms.bShape[1] * uniforms.bShape[2];
          `}
        ${t}
      }
      fn mm_write(row : i32, col : i32, valueIn : f32, globalId : vec3<u32>) {
        if (coordsInBounds2D(vec2<i32>(row, col), vec2<i32>(uniforms.dimAOuter, uniforms.dimBOuter))) {
          let batch = i32(globalId.z);
          let outCoord = vec3<i32>(batch, row, col);
          var value = valueIn;
          ${r}
          ${s}
          setOutputAtCoords(batch, row, col, value);
        }
      }
      ${Kne(this.workGroupSize)}
    `;
  }
};
function We(e) {
  let { inputs: t, attrs: n } = e, { x: s } = t, { shape: r } = n, a = w.sizeFromShape(s.shape), i = w.inferFromImplicitShape(r, a), o = w.sizeFromShape(i);
  return w.assert(a === o, () => `The new shape (${i}) has ${o} elements and the old shape (${s.shape}) has ${a} elements. The new shape and old shape must have the same number of elements.`), e.backend.incRef(s.dataId), { dataId: s.dataId, shape: i, dtype: s.dtype };
}
var Yne = { kernelName: Oo, backendName: "webgpu", kernelFunc: We };
function Pv({ a: e, b: t, transposeA: n, transposeB: s, backend: r, bias: a = null, preluActivationWeights: i = null, leakyreluAlpha: o = 0, activation: u = null }) {
  let l = e.shape.length, c = t.shape.length, p = n ? e.shape[l - 2] : e.shape[l - 1], d = s ? t.shape[c - 1] : t.shape[c - 2], h = n ? e.shape[l - 1] : e.shape[l - 2], f = s ? t.shape[c - 2] : t.shape[c - 1], m = e.shape.slice(0, -2), g = t.shape.slice(0, -2), b = w.sizeFromShape(m), y = w.sizeFromShape(g), x = Qo.assertAndGetBroadcastShape(e.shape.slice(0, -2), t.shape.slice(0, -2)).concat([h, f]);
  w.assert(p === d, () => `Error in matMul: inner shapes (${p}) and (${d}) of Tensors with shapes ${e.shape} and ${t.shape} and transposeA=${n} and transposeB=${s} must match.`);
  let k = n ? [b, p, h] : [b, h, p], I = s ? [y, f, d] : [y, d, f], $ = We({ inputs: { x: e }, backend: r, attrs: { shape: k } }), R = We({ inputs: { x: t }, backend: r, attrs: { shape: I } }), E = [$, R], P = Math.max(b, y), A = b === 1, O = y === 1, T = p % 4 === 0 && f % 4 === 0 && !n && !s, z;
  h * f <= 32 ? z = new jne([P, h, f], A, O, n, s, a, u, i) : !n && !s && (h <= 16 && (f <= 512 || d >= 2 * f) || f <= 16 && (h <= 512 || p >= 2 * h)) ? z = new Xne(k, I, [P, h, f], a, u, i) : T ? z = new Une(k, [P, h, f], K().get("WEBGPU_MATMUL_WORK_PER_THREAD"), A, O, a, u, i) : z = new Hne(k, [P, h, f], K().get("WEBGPU_MATMUL_WORK_PER_THREAD"), A, O, n, s, a, u, i);
  let W = [$, R];
  a && W.push(a), i && W.push(i);
  let q = [{ type: "int32", data: [h] }, { type: "int32", data: [f] }, { type: "int32", data: [p] }];
  u === "leakyrelu" && (q.push({ type: "float32", data: [o] }), z.uniforms += " alpha : f32,");
  let X = r.runWebGPUProgram(z, W, e.dtype, q), Y = We({ inputs: { x: X }, backend: r, attrs: { shape: x } });
  E.push(X);
  for (let Z of E)
    r.disposeData(Z.dataId);
  return Y;
}
function Qne(e) {
  let { inputs: t, backend: n, attrs: s } = e, { a: r, b: a, bias: i, preluActivationWeights: o } = t, { transposeA: u, transposeB: l, activation: c, leakyreluAlpha: p } = s;
  return Pv({ a: r, b: a, transposeA: u, transposeB: l, backend: n, bias: i, preluActivationWeights: o, leakyreluAlpha: p, activation: c });
}
var Zne = { kernelName: aa, backendName: "webgpu", kernelFunc: Qne };
var Pw = class {
  constructor(e, t, n) {
    this.variableNames = ["AReal", "AImag", "BReal", "BImag"], this.workGroupSize = [128, 1, 1], this.size = true, this.outputShape = C.assertAndGetBroadcastShape(t, n), this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = `binaryOpComplex_${e}`, this.op = e;
  }
  getUserCode() {
    return `
      fn binaryOpComplex(
          areal : f32, aimag : f32, breal : f32, bimag : f32) -> f32 {
        ${sc(this.op, false)}
      }

      ${Ue()}
        if(index < uniforms.size) {
          let areal = getARealByOutputIndex(index);
          let aimag = getAImagByOutputIndex(index);
          let breal = getBRealByOutputIndex(index);
          let bimag = getBImagByOutputIndex(index);
          setOutputAtIndex(index, binaryOpComplex(areal, aimag, breal, bimag));
        }
      }
    `;
  }
};
var Jne = class {
  constructor(e, t, n, s) {
    this.variableNames = ["A", "B"], this.size = true;
    let r = 256;
    this.workGroupSize = [r, 1, 1], this.outputShape = C.assertAndGetBroadcastShape(t, n), this.dispatchLayout = Be(this.outputShape), this.lastDimensionSize = s ? n[0] : t[0], this.lastDimensionSize < 256 ? this.workPerThread = 1 : this.lastDimensionSize < 512 ? this.workPerThread = 2 : this.workPerThread = 4, this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]), this.useSharedMemoryWithB = s, this.op = e, this.shaderKey = `binaryShared_${e}_${this.lastDimensionSize}_${this.useSharedMemoryWithB}`;
  }
  getUserCode() {
    let e = this.lastDimensionSize > 1 ? `coords[${this.outputShape.length - 1}]` : "0", t = this.useSharedMemoryWithB ? `let a = getAByOutputCoords(coords);
         let b = sharedBuf[${e}];` : `let a = sharedBuf[${e}];
         let b = getBByOutputCoords(coords);`;
    return `
        fn binaryOperation(a : f32, b : f32) -> f32 {
          ${sc(this.op, false)}
        }
        var<workgroup> sharedBuf : array<f32, ${this.lastDimensionSize}>;
        ${Ue()}

          // Fill in the shared memory buffer. Here we need a loop to make sure
          // that all data in A|B are uploaded when |sharedMemorySize| is larger
          // than work group size.
          for(var localIndex = i32(localId.x); localIndex < ${this.lastDimensionSize}; localIndex = localIndex + ${this.workGroupSize[0]}) {
            sharedBuf[localIndex] = f32(${this.useSharedMemoryWithB ? "B" : "A"}[localIndex]);
          }
          workgroupBarrier();

          for(var i = 0; i < ${this.workPerThread}; i = i + 1) {
            let flatIndex = index * ${this.workPerThread} + i;
            if(flatIndex < uniforms.size) {
              let coords = getCoordsFromIndex(flatIndex);

              ${t}
              setOutputAtIndex(flatIndex, binaryOperation(a, b));
            }
          }
        }
        `;
  }
};
var ese = class {
  constructor(e, t, n) {
    this.variableNames = ["A", "B"], this.workPerThread = 4, this.isVec4 = true, this.size = true;
    let s = 128;
    this.workGroupSize = [s, 1, 1], this.outputShape = C.assertAndGetBroadcastShape(t, n), this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]), this.op = e, this.shaderKey = `binaryVec4_${e}`;
  }
  getUserCode() {
    return `
      fn binaryOperation(a : vec4<f32>, b : vec4<f32>) -> vec4<f32> {
        ${sc(this.op, this.isVec4)}
      }
      ${Ue()}
        if (index < uniforms.size) {
          let a = getAByOutputIndex(index);
          let b = getBByOutputIndex(index);
          setOutputAtIndex(index, binaryOperation(a, b));
        }
      }
    `;
  }
};
var A2 = class {
  constructor(e, t, n) {
    this.variableNames = ["A", "B"], this.size = true;
    let s = 128;
    this.workGroupSize = [s, 1, 1], this.outputShape = C.assertAndGetBroadcastShape(t, n), this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = `binary_${e}`, this.op = e;
  }
  getUserCode() {
    return `
      fn binaryOperation(a : f32, b : f32) -> f32 {
        ${sc(this.op, false)}
      }
      ${Ue()}
        if (index < uniforms.size) {
          let a = getAByOutputIndex(index);
          let b = getBByOutputIndex(index);
          setOutputAtIndex(index, binaryOperation(a, b));
        }
      }
      `;
  }
};
function zw(e, t, n) {
  if (w.arraysEqual(t, n) && w.sizeFromShape(t) % 4 === 0)
    return new ese(e, t, n);
  let r = t.length === 1 && n.length > 1 && t[0] < 1024, a = n.length === 1 && t.length > 1 && n[0] < 1024;
  return r || a ? new Jne(e, t, n, a) : new A2(e, t, n);
}
function Wn(e) {
  let { inputs: t } = e, { x: n } = t;
  return e.backend.incRef(n.dataId), { dataId: n.dataId, shape: n.shape, dtype: n.dtype };
}
var tse = { kernelName: Wa, backendName: "webgpu", kernelFunc: Wn };
function hu(e) {
  let { inputs: t, backend: n } = e, { real: s, imag: r } = t, a = n.makeTensorInfo(s.shape, "complex64"), i = n.tensorMap.get(a.dataId), o = Wn({ inputs: { x: s }, backend: n }), u = Wn({ inputs: { x: r }, backend: n });
  return i.complexTensorInfos = { real: o, imag: u }, a;
}
var nse = { kernelName: ep, backendName: "webgpu", kernelFunc: hu };
var rc = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.size = true;
    let n = 128;
    this.workGroupSize = [n, 1, 1], this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.op = t, this.shaderKey = `unary_${t}`;
  }
  getUserCode() {
    return `
      fn unaryOperation(a : f32) -> f32 {
        ${qr(this.op, false)}
      }
      ${Ue()}
        if (index < uniforms.size) {
          let a = getAByOutputIndex(index);
          setOutputAtIndex(index, unaryOperation(a));
        }
      }
      `;
  }
};
function Kt({ opType: e, cpuKernelImpl: t, dtype: n }) {
  return ({ inputs: s, backend: r }) => {
    let { x: a } = s, i = r, o = n || a.dtype;
    if (i.shouldExecuteOnCPU([a]) && t != null) {
      let l = i.tensorMap.get(a.dataId), c = t(l.values, o);
      return i.makeTensorInfo(a.shape, o, c);
    }
    let u = new rc(a.shape, e);
    return i.runWebGPUProgram(u, [a], o);
  };
}
function mn({ opSnippet: e, cpuKernelImpl: t, supportsComplex: n = false, dtype: s }) {
  return ({ inputs: r, backend: a }) => {
    let { a: i, b: o } = r, u = a;
    if (n && i.dtype === "complex64") {
      let p = u.tensorMap.get(i.dataId), d = u.tensorMap.get(o.dataId), h, f;
      if (e !== 0)
        [h, f] = [[p.complexTensorInfos.real, d.complexTensorInfos.real], [p.complexTensorInfos.imag, d.complexTensorInfos.imag]].map((g) => {
          let [b, y] = g, v = { dataId: b.dataId, dtype: b.dtype, shape: i.shape }, x = { dataId: y.dataId, dtype: y.dtype, shape: o.shape }, k = zw(e, i.shape, o.shape);
          return u.runWebGPUProgram(k, [v, x], cn(b.dtype, y.dtype));
        });
      else {
        let g = new Pw(17, i.shape, o.shape), b = new Pw(18, i.shape, o.shape), y = [{ dataId: p.complexTensorInfos.real.dataId, dtype: p.complexTensorInfos.real.dtype, shape: i.shape }, { dataId: p.complexTensorInfos.imag.dataId, dtype: p.complexTensorInfos.imag.dtype, shape: i.shape }, { dataId: d.complexTensorInfos.real.dataId, dtype: d.complexTensorInfos.real.dtype, shape: o.shape }, { dataId: d.complexTensorInfos.imag.dataId, dtype: d.complexTensorInfos.imag.dtype, shape: o.shape }];
        h = u.runWebGPUProgram(g, y, "float32"), f = u.runWebGPUProgram(b, y, "float32");
      }
      let m = hu({ inputs: { real: h, imag: f }, backend: u });
      return u.disposeData(h.dataId), u.disposeData(f.dataId), m;
    }
    let l = s || cn(i.dtype, o.dtype);
    if ((i.dtype === "string" || o.dtype === "string" || u.shouldExecuteOnCPU([i, o])) && t != null) {
      let p = u.tensorMap.get(i.dataId).values, d = u.tensorMap.get(o.dataId).values, h = i.dtype === "string" ? C.fromUint8ToStringArray(p) : p, f = i.dtype === "string" ? C.fromUint8ToStringArray(d) : d, [m, g] = t(i.shape, o.shape, h, f, l);
      return u.makeTensorInfo(g, l, m);
    }
    let c = zw(e, i.shape, o.shape);
    return u.runWebGPUProgram(c, [i, o], l);
  };
}
var { addImpl: sse, ceilImpl: rse, concatImpl: ase, equalImpl: ise, expImpl: ose, expm1Impl: use, floorImpl: lse, gatherNdImpl: cse, gatherV2Impl: dse, greaterEqualImpl: pse, greaterImpl: hse, lessEqualImpl: fse, lessImpl: mse, logImpl: gse, maxImpl: bse, maximumImpl: yse, minimumImpl: vse, multiplyImpl: xse, negImpl: wse, notEqualImpl: kse, prodImpl: Sse, rangeImpl: Ise, rsqrtImpl: Cse, scatterImpl: Nse, simpleAbsImpl: Tse, sliceImpl: $se, stridedSliceImpl: _se, stringNGramsImpl: Ase, subImpl: Ese, tileImpl: Rse, topKImpl: Dse, transposeImpl: Fse, uniqueImpl: ghe } = iv;
var Ose = Kt({ opType: 0, cpuKernelImpl: Tse });
var Pse = { kernelName: po, backendName: "webgpu", kernelFunc: Ose };
var zse = mn({ opSnippet: 1, cpuKernelImpl: sse, supportsComplex: true });
var Mse = { kernelName: Sr, backendName: "webgpu", kernelFunc: zse };
var Lse = class {
  constructor(e) {
    this.workPerThread = 4, this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e[0], this.variableNames = e.map((t, n) => `T${n}`), this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]), this.shaderKey = "addN";
  }
  getUserCode() {
    let e = [];
    this.variableNames.forEach((s) => {
      e.push(`let v${s} = get${s}ByOutputCoords(coords);`);
    });
    let t = this.variableNames.map((s) => `v${s}`).join(" + ");
    return `
      ${Ue()}
        for (var i = 0; i < ${this.workPerThread}; i = i + 1) {
          let flatIndex = index * ${this.workPerThread} + i;
          if (flatIndex < uniforms.size) {
            let coords = getCoordsFromIndex(flatIndex);
            ${e.join(`
        `)}
            setOutputAtIndex(flatIndex, ${t});
          }
        }
      }
    `;
  }
};
function Bse(e) {
  let { inputs: t, backend: n } = e, s = t;
  if (s.length === 1)
    return Wn({ inputs: { x: s[0] }, backend: n });
  let r = s.map((o) => o.dtype).reduce((o, u) => cn(o, u)), a = s.map((o) => o.shape), i = new Lse(a);
  return n.runWebGPUProgram(i, s, r);
}
var Vse = { kernelName: Sa, backendName: "webgpu", kernelFunc: Bse };
var E2 = class {
  constructor(e, t, n) {
    this.workGroupSize = [64, 1, 1], this.variableNames = ["x"], this.uniforms = "infinityValue : f32,", this.size = true;
    let s = [t];
    C.assertAxesAreInnerMostDims("arg" + n.charAt(0).toUpperCase() + n.slice(1), s, e.length), this.op = n === "min" ? "<" : ">";
    let [r] = C.computeOutAndReduceShapes(e, s);
    this.outputShape = r.length === 0 ? [1] : r, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, [1, 1, 1]), this.inputShape = e, this.shaderKey = `argMinMax${this.op}`;
  }
  getUserCode() {
    let e = `
      var<workgroup> xBestIndices : array<i32, ${this.workGroupSize[0]}>;
      var<workgroup> xBestValues : array<f32, ${this.workGroupSize[0]}>;
    `, t = () => this.inputShape.length === 1 ? "uniforms.xShape" : `uniforms.xShape.${pr(this.inputShape.length - 1)}`, n = () => {
      let r = "";
      if (this.outputShape.length === 1)
        this.inputShape.length !== 1 && (r += "outputCoords,");
      else
        for (let a = 0; a < this.outputShape.length; a++)
          r += `outputCoords.${pr(a)},`;
      return r;
    };
    return `
      fn DIV_CEIL(a : u32, b : u32) -> u32 {
        return ((a - 1u) / b + 1u);
      }

      ${e}

      ${Ue()}
        let outputIndex = index / i32(workGroupSizeX);
        let reduceLength = ${t()};

        var bestIndex = i32(localId.x);
        var bestValue = uniforms.infinityValue;
        let outputCoords = getCoordsFromIndex(outputIndex);
        for (var k = i32(localId.x); k < reduceLength && outputIndex < uniforms.size;
            k = k + i32(workGroupSizeX)) {
          let candidate = getX(${n()} k);
          if (!isnan(candidate) && candidate ${this.op} bestValue) {
            bestValue = candidate;
            bestIndex = k;
          }
        }
        xBestValues[localId.x] = bestValue;
        xBestIndices[localId.x] = bestIndex;
        workgroupBarrier();

        var reduceSize = min(u32(reduceLength), workGroupSizeX);
        for (var currentSize = reduceSize / 2u; reduceSize > 1u;
            currentSize = reduceSize / 2u) {
          let interval = DIV_CEIL(reduceSize, 2u);
          if (localId.x < currentSize) {
            let candidate = xBestValues[localId.x + interval];
            if (candidate ${this.op} bestValue) {
              bestValue = candidate;
              xBestValues[localId.x] = bestValue;
              xBestIndices[localId.x] = xBestIndices[localId.x + interval];
            }
          }
          reduceSize = interval;
          workgroupBarrier();
        }

        if (localId.x == 0u && outputIndex < uniforms.size) {
          setOutputAtIndexI32(outputIndex, xBestIndices[localId.x]);
        }
      }
    `;
  }
};
var Wse = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.workGroupSize = [16, 16, 1];
    let n = new Array(e.length);
    for (let s = 0; s < n.length; s++)
      n[s] = e[t[s]];
    this.outputShape = n, this.dispatchLayout = { x: [0], y: [1] }, this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [1, 1, 1]), this.shaderKey = "transposeShared";
  }
  getUserCode() {
    return `
      let TILE_DIM = ${this.workGroupSize[0]};
      var<workgroup> tile : array<array<f32, ${this.workGroupSize[0] + 1}>, ${this.workGroupSize[0]}>;
      ${Ev()}
      fn main(@builtin(local_invocation_id) localId : vec3<u32>,
              @builtin(workgroup_id) workgroupId : vec3<u32>) {
        var x = i32(workgroupId.x) * TILE_DIM + i32(localId.x);
        var y = i32(workgroupId.y) * TILE_DIM + i32(localId.y);
        let width = uniforms.outShape[0];
        let height = uniforms.outShape[1];
        if (x < width && y < height) {
          tile[localId.y][localId.x] = A[y * width + x];
        }
        workgroupBarrier();

        x = i32(workgroupId.y) * TILE_DIM + i32(localId.x);
        y = i32(workgroupId.x) * TILE_DIM + i32(localId.y);
        if (x < height && y < width) {
          setOutputAtIndex((y * height + x), tile[localId.x]
            [localId.y]);
        }
      }
    `;
  }
};
var Use = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.workPerThread = 4, this.workGroupSize = [64, 1, 1], this.size = true;
    let n = new Array(e.length);
    for (let s = 0; s < n.length; s++)
      n[s] = e[t[s]];
    this.outputShape = n, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]), this.newDim = t, this.shaderKey = `transpose_${t}`;
  }
  getUserCode() {
    let e = Ut(this.outputShape.length), t = Gse(this.newDim);
    return `
      ${Ue()}

        for(var i = 0; i < ${this.workPerThread}; i = i + 1) {
          let flatIndex = index * ${this.workPerThread} + i;
          if(flatIndex < uniforms.size) {
            let resRC = getCoordsFromIndex(flatIndex);
            setOutputAtIndex(flatIndex, A[getIndexFromCoords${this.outputShape.length}D(
              ${e}(${t}), uniforms.aShape)]);
          }
        }
      }
    `;
  }
};
function Gse(e) {
  let t = e.length;
  if (t > 6)
    throw Error(`Transpose for rank ${t} is not yet supported`);
  let n = new Array(t);
  for (let s = 0; s < e.length; s++)
    n[e[s]] = `resRC.${pr(s)}`;
  return n.join();
}
function Ks(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { perm: a } = s, i = n, o = r.shape.length, u = new Array(o);
  for (let c = 0; c < u.length; c++)
    u[c] = r.shape[a[c]];
  if (n.shouldExecuteOnCPU([r])) {
    let p = i.tensorMap.get(r.dataId).values, d = Fse(p, r.shape, r.dtype, a, u);
    return n.makeTensorInfo(u, r.dtype, d);
  }
  if (r.shape.length === 2 && w.arraysEqual(a, [1, 0])) {
    let c = new Wse(r.shape, a);
    return i.runWebGPUProgram(c, [r], r.dtype);
  }
  let l = new Use(r.shape, a);
  return i.runWebGPUProgram(l, [r], r.dtype);
}
var Hse = { kernelName: mi, backendName: "webgpu", kernelFunc: Ks };
function qse(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a } = s, i = w.parseAxisParam(a, r.shape), o = C.getAxesPermutation(i, r.shape.length), u = r, l = [];
  o != null && (u = Ks({ inputs: { x: r }, backend: n, attrs: { perm: o } }), l.push(u), i = C.getInnerMostAxes(i.length, u.shape.length)), C.assertAxesAreInnerMostDims("argMax", [i[0]], u.shape.length);
  let c = new E2(u.shape, i[0], "max"), p = [{ type: "float32", data: [Number.NEGATIVE_INFINITY] }], d = n.runWebGPUProgram(c, [u], "int32", p);
  return l.forEach((h) => n.disposeData(h.dataId)), d;
}
var jse = { kernelName: Ia, backendName: "webgpu", kernelFunc: qse };
function Kse(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a } = s, i = w.parseAxisParam(a, r.shape), o = C.getAxesPermutation(i, r.shape.length), u = r, l = [];
  o != null && (u = Ks({ inputs: { x: r }, backend: n, attrs: { perm: o } }), l.push(u), i = C.getInnerMostAxes(i.length, u.shape.length)), C.assertAxesAreInnerMostDims("argMin", [i[0]], u.shape.length);
  let c = new E2(u.shape, i[0], "min"), p = [{ type: "float32", data: [Number.POSITIVE_INFINITY] }], d = n.runWebGPUProgram(c, [u], "int32", p);
  return l.forEach((h) => n.disposeData(h.dataId)), d;
}
var Xse = { kernelName: dl, backendName: "webgpu", kernelFunc: Kse };
var R2 = class {
  constructor(e, t) {
    this.variableNames = ["x"], this.uniforms = "stride : vec2<i32>, pad : vec2<i32>, dilation : vec2<i32>, convDims : vec2<i32>, filterDims : vec2<i32>,", this.workGroupSize = [128, 1, 1], this.size = true, this.outputShape = e.outShape, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = `pool2D_${t}`, this.poolType = t;
  }
  getUserCode() {
    let e = "resultValue = max(value, resultValue);";
    this.poolType === "avg" && (e = "resultValue = resultValue + value; count = count + 1.0;");
    let t = "resultValue";
    return this.poolType === "avg" && (t = "resultValue / count"), `
      ${Ue()}
      if (index < uniforms.size) {
        let coords = getCoordsFromIndex(index);
          let batch = coords[0];
          let xRCCorner = vec2<i32>(coords.yz) * uniforms.stride - uniforms.pad;
          let xRCorner = xRCCorner.x;
          let xCCorner = xRCCorner.y;

          var resultValue = ${this.poolType === "avg" ? "0.0" : "-1.0 / pow(10.0, -20.0)"};
          var count = 0.0;

          for (var wR = 0; wR < uniforms.filterDims.x; wR = wR + uniforms.dilation.x) {
            let xR = xRCorner + wR;

            if (xR < 0 || xR >= uniforms.convDims.x) {
              continue;
            }

            for (var wC = 0; wC < uniforms.filterDims.y; wC = wC + uniforms.dilation.y) {
              let xC = xCCorner + wC;
              if (xC < 0 || xC >= uniforms.convDims.y) {
                continue;
              }

              let value = getX(batch, xR, xC, coords[3]);
              ${e}
            }
          }

          setOutputAtIndex(index, ${t});
        }
      }
    `;
  }
};
var D2 = class {
  constructor(e) {
    this.variableNames = ["x"], this.uniforms = "stride : vec2<i32>,", this.workGroupSize = [256, 1, 1], this.size = true, this.outputShape = e.outShape, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = "poolWithFilterSizeEqualsOne";
  }
  getUserCode() {
    return `
      ${Ue()}
        if (index < uniforms.size) {
          let coords = getCoordsFromIndex(index);
          let batch = coords[0];
          let d = coords[3];

          let xRCCorner = coords.yz * uniforms.stride;
          let xRCorner = xRCCorner.x;
          let xCCorner = xRCCorner.y;

          let value = getX(batch, xRCorner, xCCorner, d);
          setOutputAtIndex(index, value);
        }
      }
    `;
  }
};
function Yse(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { filterSize: a, strides: i, pad: o, dimRoundingMode: u } = s, l = 1, c = C.computePool2DInfo(r.shape, a, i, l, o, u);
  if (c.filterWidth === 1 && c.filterHeight === 1 && w.arraysEqual(c.inShape, c.outShape))
    return Wn({ inputs: { x: r }, backend: n });
  let p, d = [{ type: "int32", data: [c.strideHeight, c.strideWidth] }];
  return c.filterHeight === 1 && c.filterWidth === 1 ? p = new D2(c) : (p = new R2(c, "avg"), d.push({ type: "int32", data: [c.padInfo.top, c.padInfo.left] }, { type: "int32", data: [c.dilationHeight, c.dilationWidth] }, { type: "int32", data: [c.inHeight, c.inWidth] }, { type: "int32", data: [c.effectiveFilterHeight, c.effectiveFilterWidth] })), n.runWebGPUProgram(p, [r], r.dtype, d);
}
var Qse = { kernelName: Ca, backendName: "webgpu", kernelFunc: Yse };
function Zse(e) {
  let { inputs: t, backend: n, attrs: s } = e, { a: r, b: a } = t, { transposeA: i, transposeB: o } = s;
  return Pv({ a: r, b: a, transposeA: i, transposeB: o, backend: n });
}
var Jse = { kernelName: Na, backendName: "webgpu", kernelFunc: Zse };
var ere = class {
  constructor(e, t) {
    this.variableNames = ["source"], this.workPerThread = 1, this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = t, this.rank = t.length, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]), this.start = e, this.uniforms = `start : ${Ut(e.length)}, `, this.shaderKey = "slice";
  }
  getUserCode() {
    let e = Ut(this.rank), t = tre(this.rank), n;
    return this.start.length === 1 ? n = this.outputShape.map((r, a) => "sourceLoc = uniforms.start + coords;") : n = this.outputShape.map((r, a) => `sourceLoc.${eg[a]} = uniforms.start[${a}] + coords.${eg[a]};`), `
      ${Ue()}
        if (index < uniforms.size) {
          var sourceLoc : ${e};
          let coords = getCoordsFromIndex(index);
          ${n.join(`
`)}
          setOutputAtIndex(index, getSource(${t}));
        }
      }
    `;
  }
};
var eg = ["x", "y", "z", "w", "u", "v"];
function tre(e) {
  if (e === 1)
    return "sourceLoc";
  if (e <= 6)
    return eg.slice(0, e).map((t) => `sourceLoc.${t}`).join(",");
  throw Error(`Slicing for rank ${e} is not yet supported`);
}
function fu(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { begin: a, size: i } = s, [o, u] = wt.parseSliceParams(r, a, i);
  if (wt.assertParamsValid(r, o, u), n.shouldExecuteOnCPU([r]) || r.dtype === "string") {
    let p = n.tensorMap.get(r.dataId), d = $se(p.values, o, u, r.shape, r.dtype);
    return n.makeTensorInfo(u, r.dtype, d);
  }
  if (w.sizeFromShape(u) === 0)
    return n.makeTensorInfo(u, r.dtype, []);
  let l = new ere(o, u), c = [{ type: "int32", data: o }];
  return n.runWebGPUProgram(l, [r], r.dtype, c);
}
var nre = { kernelName: Bo, backendName: "webgpu", kernelFunc: fu };
var sre = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockShape: a, crops: i } = s;
  w.assert(r.shape.length <= 4, () => "batchToSpaceND for rank > 4 with a WebGPU backend not implemented yet");
  let o = a.reduce((y, v) => y * v), u = C.getReshaped(r.shape, a, o), l = C.getPermuted(u.length, a.length), c = C.getReshapedPermuted(r.shape, a, o), p = C.getSliceBeginCoords(i, a.length), d = C.getSliceSize(c, i, a.length), h = [], f = We({ inputs: { x: r }, backend: n, attrs: { shape: u } }), m = Ks({ inputs: { x: f }, backend: n, attrs: { perm: l } }), g = We({ inputs: { x: m }, backend: n, attrs: { shape: c } }), b = fu({ inputs: { x: g }, backend: n, attrs: { begin: p, size: d } });
  return h.push(f), h.push(m), h.push(g), h.forEach((y) => n.disposeData(y.dataId)), b;
};
var rre = { kernelName: ho, backendName: "webgpu", kernelFunc: sre };
var F2 = mn({ opSnippet: 10, dtype: "bool", cpuKernelImpl: kse });
var are = { kernelName: _o, backendName: "webgpu", kernelFunc: F2 };
function ac(e) {
  let { inputs: t, backend: n } = e, { input: s } = t, r = n.tensorMap.get(s.dataId);
  return Wn({ inputs: { x: r.complexTensorInfos.real }, backend: n });
}
var ire = { kernelName: lp, backendName: "webgpu", kernelFunc: ac };
function ore(e, t) {
  let n = new rc(e.shape, 22), s = t.runWebGPUProgram(n, [e], "int32");
  return { dataId: s.dataId, shape: s.shape, dtype: s.dtype };
}
function tg(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { dtype: a } = s;
  if (a === "complex64") {
    if (r.dtype === "complex64")
      return Wn({ inputs: { x: r }, backend: n });
    let i = $t(r.shape), o = tg({ inputs: { x: r }, backend: n, attrs: { dtype: "float32" } }), u = hu({ inputs: { real: o, imag: i }, backend: n });
    return i.dispose(), n.disposeData(o.dataId), u;
  }
  if (r.dtype === "complex64") {
    let i = ac({ inputs: { input: r }, backend: n }), o = tg({ inputs: { x: i }, backend: n, attrs: { dtype: a } });
    return n.disposeData(i.dataId), o;
  }
  if (!w.hasEncodingLoss(r.dtype, a)) {
    let i = Wn({ inputs: { x: r }, backend: n });
    return { dataId: i.dataId, shape: i.shape, dtype: a };
  }
  if (a === "int32")
    return ore(r, n);
  if (a === "bool") {
    let i = n.makeTensorInfo([], "bool", w.getTypedArrayFromDType("bool", 1)), u = F2({ inputs: { a: r, b: i }, backend: n });
    return n.disposeData(i.dataId), u;
  }
  throw new Error(`Error in Cast: failed to cast ${r.dtype} to ${a}`);
}
var ure = { kernelName: Ta, backendName: "webgpu", kernelFunc: tg };
var lre = Kt({ opType: 1, cpuKernelImpl: rse });
var cre = { kernelName: $a, backendName: "webgpu", kernelFunc: lre };
var dre = class {
  constructor(e) {
    this.variableNames = ["A"], this.uniforms = "minVal : f32, maxVal : f32,", this.workPerThread = 4, this.workGroupSize = [64, 1, 1], this.isVec4 = true, this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]), this.shaderKey = "clipVec4";
  }
  getUserCode() {
    return `
      ${Ue()}
        if(index < uniforms.size) {
          let value = getAByOutputIndex(index);
          var clampedValue : vec4<f32>;
          for (var i = 0; i < 4; i = i + 1) {
            if (isnan(value[i])) {
              clampedValue[i] = value[i];
            } else {
              clampedValue[i] = clamp(value[i], uniforms.minVal, uniforms.maxVal);
            }
          }

          setOutputAtIndex(index, clampedValue);
        }
      }
    `;
  }
};
var pre = class {
  constructor(e) {
    this.variableNames = ["A"], this.uniforms = "minVal : f32, maxVal : f32,", this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = "clip";
  }
  getUserCode() {
    return `
      ${Ue()}
        if(index < uniforms.size) {
          let value = getAByOutputIndex(index);
          if (isnan(value)) {
            setOutputAtIndex(index, value);
            return;
          }
          setOutputAtIndex(index, clamp(value, uniforms.minVal, uniforms.maxVal));
        }
      }
    `;
  }
};
function hre(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { clipValueMin: a, clipValueMax: i } = s, o, u = [{ type: "float32", data: [a] }, { type: "float32", data: [i] }];
  return w.sizeFromShape(r.shape) % 4 === 0 ? o = new dre(r.shape) : o = new pre(r.shape), n.runWebGPUProgram(o, [r], r.dtype, u);
}
var fre = { kernelName: Ir, backendName: "webgpu", kernelFunc: hre };
var mre = class {
  constructor(e) {
    this.uniforms = "", this.workPerThread = 4, this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = C.computeOutShape(e, 1), this.variableNames = e.map((t, n) => `T${n}`), this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]), this.offsetLength = e.length - 1;
    for (let t = 0; t < this.offsetLength; t++)
      this.uniforms += `offset${t} : i32,`;
    this.shaderKey = "concat";
  }
  getUserCode() {
    let e = [];
    if (this.offsetLength > 0) {
      e.push("if (yC < uniforms.offset0){ setOutputAtCoords(coords.x, coords.y, getT0(yR, yC)); }");
      for (let r = 1; r < this.offsetLength; r++)
        e.push(`else if (yC < uniforms.offset${[r]}){ setOutputAtCoords(coords.x, coords.y, getT${r}(yR, yC - uniforms.offset${r - 1})); }`);
      let n = this.offsetLength, s = this.offsetLength - 1;
      e.push(`else { setOutputAtCoords(coords.x, coords.y, getT${n}(yR, yC - uniforms.offset${s})); }`);
    } else
      e.push("setOutputAtCoords(coords.x, coords.y, getT0(yR, yC));");
    return `
      ${Ue()}
        for(var i = 0; i < ${this.workPerThread}; i = i + 1) {
          let flatIndex = index * ${this.workPerThread} + i;
          if(flatIndex < uniforms.size) {
            let coords = getCoordsFromIndex(flatIndex);
            let yR = coords.x;
            let yC = coords.y;

            ${e.join(`
        `)}
          }
        }
      }
    `;
  }
};
function ah(e) {
  let { inputs: t, backend: n } = e, { input: s } = t, r = n.tensorMap.get(s.dataId);
  return Wn({ inputs: { x: r.complexTensorInfos.imag }, backend: n });
}
var gre = { kernelName: ap, backendName: "webgpu", kernelFunc: ah };
function ng(e, t, n) {
  let s = e[0].dtype;
  if (s === "complex64") {
    let h = e.map((y) => ac({ inputs: { input: y }, backend: n })), f = e.map((y) => ah({ inputs: { input: y }, backend: n })), m = ng(h, t, n), g = ng(f, t, n), b = hu({ inputs: { real: m, imag: g }, backend: n });
    return h.forEach((y) => n.disposeData(y.dataId)), f.forEach((y) => n.disposeData(y.dataId)), n.disposeData(m.dataId), n.disposeData(g.dataId), b;
  }
  let r = n.shouldExecuteOnCPU(e);
  if (s === "string" && (r = true), r) {
    let h = e.map((x) => {
      let k = w.sizeFromShape(x.shape.slice(t));
      return We({ inputs: { x }, backend: n, attrs: { shape: [-1, k] } });
    }), f = h.map((x) => ({ vals: n.readSync(x.dataId), shape: x.shape })), m = C.computeOutShape(h.map((x) => x.shape), 1), g = h[0].shape[0] === 1, b = ase(f, m, s, g), y = C.computeOutShape(e.map((x) => x.shape), t), v = n.makeTensorInfo(y, s, b);
    return h.forEach((x) => n.disposeData(x.dataId)), v;
  }
  let { tensors2D: a, outShape: i } = bre(e, t, n), o = a.map((h) => h.shape), u = new mre(o), l = [], c = new Array(o.length - 1);
  if (c.length > 0) {
    c[0] = o[0][1], l.push({ type: "int32", data: [c[0]] });
    for (let h = 1; h < c.length; h++)
      c[h] = c[h - 1] + o[h][1], l.push({ type: "int32", data: [c[h]] });
  }
  let p = n.runWebGPUProgram(u, a, a[0].dtype, l);
  a.forEach((h) => n.disposeData(h.dataId));
  let d = We({ inputs: { x: p }, backend: n, attrs: { shape: i } });
  return n.disposeData(p.dataId), d;
}
function bre(e, t, n) {
  let s = C.computeOutShape(e.map((a) => a.shape), t);
  return { tensors2D: e.map((a) => We({ inputs: { x: a }, backend: n, attrs: { shape: [w.sizeFromShape(a.shape.slice(0, t)), w.sizeFromShape(a.shape.slice(t))] } })), outShape: s };
}
function O2(e) {
  let { inputs: t, backend: n, attrs: s } = e, { axis: r } = s, a = w.parseAxisParam(r, t[0].shape)[0], i = C.computeOutShape(t.map((l) => l.shape), a);
  if (w.sizeFromShape(i) === 0)
    return n.makeTensorInfo(i, t[0].dtype, []);
  let o = t.filter((l) => w.sizeFromShape(l.shape) > 0);
  if (o.length === 1)
    return Wn({ inputs: { x: o[0] }, backend: n });
  let u = o.map((l) => l.shape);
  return C.assertParamsConsistent(u, a), ng(o, a, n);
}
var yre = { kernelName: fo, backendName: "webgpu", kernelFunc: O2 };
var vre = class {
  constructor(e, t = false, n = null, s = false) {
    this.variableNames = ["x", "W"], this.uniforms = `filterDims : vec2<i32>, pad : vec2<i32>, stride : vec2<i32>, dilation : vec2<i32>,
      dimAOuter : i32, dimBOuter : i32, dimInner : i32,`, this.workGroupSize = [8, 8, 1], this.isVec4 = true, this.outputShape = e.outShape, w.assert(e.dataFormat === "channelsLast", () => "TODO: NCHW is unimplemented"), this.dispatchLayout = { x: [3], y: [1, 2], z: [0] }, this.outputShape[1] === 1 ? this.elementsPerThread = [4, 1, 1] : this.elementsPerThread = [4, 4, 1], this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, this.elementsPerThread), this.convInfo = e, this.addBias = t, this.activation = n, this.hasPreluActivationWeights = s, this.addBias && this.variableNames.push("bias"), this.hasPreluActivationWeights && this.variableNames.push("preluActivationWeights"), this.tileAOuter = this.outputShape[1] === 1 ? 1 : this.workGroupSize[1] * this.elementsPerThread[1], this.tileBOuter = this.workGroupSize[0] * this.elementsPerThread[0], this.tileInner = this.tileBOuter, [this.fitA, this.fitB] = this.getShapeFit(), this.remainder = this.convInfo.inChannels % 4 === 0, this.shaderKey = `conv2DMMVec4_${this.activation}_${this.fitA}_${this.fitB}_${this.elementsPerThread}_${this.remainder}`;
  }
  getShapeFit() {
    let e = [this.tileAOuter, this.tileInner], t = [this.tileInner, this.tileBOuter], n = this.outputShape[1] * this.outputShape[2], s = this.outputShape[3], r = this.convInfo.filterHeight * this.convInfo.filterWidth * this.convInfo.inChannels;
    return [js(e, [n, r]), js(t, [r, s])];
  }
  getSampleAWithRemainder(e) {
    return `let flatIndex${e} = getIndexFromCoords4D(coord, uniforms.xShape);
    let divBy4Remainder${e} = flatIndex${e} % 4;
    let divBy4Index${e} = flatIndex${e} / 4;
    let curData${e} = x[divBy4Index${e}];
    if (divBy4Remainder${e} == 0) {
      temp = curData${e};
    } else {
      // TODO: This could end up being a redundant load with another one in
      // the same shader invocation. Perhaps there's an opportunity for
      // optimization
      let nextData${e} = x[divBy4Index${e} + 1];
      if (divBy4Remainder${e} == 1) {
        temp = vec4<f32>(curData${e}.yzw, nextData${e}.x);
      } else if (divBy4Remainder${e} == 2) {
        temp = vec4<f32>(curData${e}.zw, nextData${e}.xy);
      } else if (divBy4Remainder${e} == 3) {
        temp = vec4<f32>(curData${e}.w, nextData${e}.xyz);
      }
    }
    `;
  }
  getUserCode() {
    let e = _2(this.elementsPerThread, this.tileAOuter, this.tileBOuter, this.tileInner), n = `let outRow = r / uniforms.outShape[2];
        let outCol = r % uniforms.outShape[2];
        let WRow = c / (uniforms.filterDims[1] * uniforms.xShape[3]);
        let WCol = c / uniforms.xShape[3] % uniforms.filterDims[1];
        let inChCoord = c % uniforms.xShape[3];
        var coord = vec4<i32>(
            batch,
            outRow * uniforms.stride[0] + uniforms.dilation[0] * WRow - uniforms.pad[0],
            outCol * uniforms.stride[1] + uniforms.dilation[1] * WCol - uniforms.pad[1],
            inChCoord);
        var resData = vec4<f32>(0.0);
        ${this.remainder ? `// The bounds checking is always needed since we use it to pad zero for
          // the 'same' padding type.
          if (coordsInBounds4D(coord, uniforms.xShape)) {
            resData = x[getIndexFromCoords4D(coord, uniforms.xShape) / 4];
          } else {
            resData = vec4<f32>(0.0); }` : `var temp = vec4<f32>(0.0);
          ${this.getSampleAWithRemainder(1)}
          resData = temp;
          if (WCol == (uniforms.filterDims[1] - 1)) {
            coord = vec4<i32>(
              coord.x, coord.y + 1, coord.z + 1 - uniforms.filterDims[1], 0);
              ${this.getSampleAWithRemainder(2)}
            if (inChCoord == 0) {
              resData = vec4<f32>(resData.xyz, temp.x);
            } else if (inChCoord == 1) {
              resData = vec4<f32>(resData.xy, temp.xy);
            } else {
              resData = vec4<f32>(resData.x, temp.xyz);
            }
          }
          `}
        return resData;`, s = this.fitA ? `${n}` : `if (r < uniforms.dimAOuter && c < uniforms.dimInner) {
          ${n}
         }
         return vec4<f32>(0.0);
        `, r = this.fitB ? "return W[row * uniforms.dimBOuter / 4 + col];" : `if(coordsInBounds2D(vec2<i32>(row, col * 4), vec2<i32>(uniforms.dimInner, uniforms.dimBOuter))) {
           return W[row * uniforms.dimBOuter / 4 + col];
         }
         return vec4<f32>(0.0);
        `, a = "", i = "";
    if (this.activation) {
      let l = Fr(this.activation, this.isVec4);
      this.hasPreluActivationWeights ? a = `fn activation(a : vec4<f32>, outCoord : vec4<i32>) -> vec4<f32> {
          let b = getPreluActivationWeightsByOutputCoords(outCoord);
          ${l}
        }` : a = `
        fn activation(a : vec4<f32>, outCoord : vec4<i32>) -> vec4<f32> {
          ${l}
        }`, i = "value = activation(value, outCoord);";
    }
    let o = this.addBias ? "value = value + getBiasByOutputCoords(outCoord);" : "";
    return `
        ${a}
        fn mm_readA(row : i32, col : i32, globalId : vec3<u32>) -> vec4<f32> {
          let r = row;
          let c = col * 4;
          var batch = i32(globalId.z);
          ${s}
        }

        fn mm_readB(row : i32, col : i32, globalId : vec3<u32>) -> vec4<f32> {
          ${r}
        }

        fn mm_write(row : i32, col : i32, valueInput : vec4<f32>, globalId : vec3<u32>) {
          var batch = i32(globalId.z);
          var value = valueInput;
          if (row < uniforms.dimAOuter && col * 4 < uniforms.dimBOuter)
          {
            let outCoord = vec4<i32>(
              batch,
              row / uniforms.outShape[2],
              row % uniforms.outShape[2],
              col * 4);
            ${o}
            ${i}
            setOutputAtCoords(outCoord[0], outCoord[1], outCoord[2], outCoord[3],
              value);
          }
        }
        ${e}
      `;
  }
};
var xre = class {
  constructor(e, t = false, n = null, s = false) {
    this.variableNames = ["x", "W"], this.uniforms = "filterDims : vec2<i32>, pad : vec2<i32>, stride : vec2<i32>, dilation : vec2<i32>, dimAOuter : i32, dimBOuter : i32, dimInner : i32,", this.outputShape = e.outShape, this.isChannelsLast = e.dataFormat === "channelsLast", this.dispatchLayout = this.isChannelsLast ? { x: [3], y: [1, 2], z: [0] } : { x: [1], y: [2, 3], z: [0] }, this.workGroupSize = Rv(this.dispatchLayout, this.outputShape), this.elementsPerThread = Dv(this.dispatchLayout, this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, this.elementsPerThread), t && this.variableNames.push("bias"), s && this.variableNames.push("preluActivationWeights"), this.convInfo = e, this.addBias = t, this.activation = n, this.hasPreluActivationWeights = s, [this.fitA, this.fitB] = this.getShapeFit(), this.shaderKey = `conv2DMM_${this.elementsPerThread}_${this.activation}_${this.fitA}_${this.fitB}_${this.isChannelsLast}`;
  }
  getShapeFit() {
    let e = this.workGroupSize[1] * this.elementsPerThread[1], t = this.workGroupSize[0] * this.elementsPerThread[0], n = e > t ? e : t;
    w.assert(n % this.workGroupSize[0] === 0 && n % this.workGroupSize[1] === 0, () => "tileInner must be multiple of workgroupsize.x and workgroupsize.y");
    let s = [e, n], r = [n, t], a = this.convInfo.outHeight * this.convInfo.outWidth, i = this.convInfo.outChannels, o = this.convInfo.filterHeight * this.convInfo.filterWidth * this.convInfo.inChannels;
    return [js(s, [a, o]), js(r, [o, i])];
  }
  getUserCode() {
    let e = this.isChannelsLast ? `
    let coord = vec4<i32>(batch, xRow, xCol, col % inChannels);
    ` : `
    let coord = vec4<i32>(batch, col % inChannels, xRow, xCol);
    `, t = this.isChannelsLast ? `
    let outCoord = vec4<i32>(
      batch,
      row / outWidth,
      row % outWidth,
      col);
    ` : `
    let outCoord = vec4<i32>(
      batch,
      col,
      row / outWidth,
      row % outWidth);
    `, n = Ov(this.elementsPerThread, this.workGroupSize), s = `
    let inChannels = uniforms.wShape[2];
    let outWidth = ${this.isChannelsLast ? "uniforms.outShape[2]" : "uniforms.outShape[3]"};
    let outRow = row / outWidth;
    let outCol = row % outWidth;

    let WRow = col / (uniforms.filterDims[1] * inChannels);
    let WCol = col / inChannels % uniforms.filterDims[1];
    let xRow = outRow * uniforms.stride[0] + uniforms.dilation[0] * WRow - uniforms.pad[0];
    let xCol = outCol * uniforms.stride[1] + uniforms.dilation[1] * WCol - uniforms.pad[1];
    ${e}
    // The bounds checking is always needed since we use it to pad zero for the
    // 'same' padding type.
    if(coordsInBounds4D(coord, uniforms.xShape)) {
      return x[getIndexFromCoords4D(coord, uniforms.xShape)];
    }
    return 0.0;`, r = this.fitA ? `${s}` : `if (row < uniforms.dimAOuter && col < uniforms.dimInner) {
      ${s}
    }
    return 0.0;
    `, a = this.fitB ? "return W[row * uniforms.dimBOuter + col];" : `if(coordsInBounds2D(vec2<i32>(row, col), vec2<i32>(uniforms.dimInner, uniforms.dimBOuter))) {
           return W[row * uniforms.dimBOuter + col];
	 }
	 return 0.0;
	 `, i = "", o = "";
    if (this.activation) {
      let c = Fr(this.activation, false);
      this.hasPreluActivationWeights ? i = `fn activation(a: f32, outCoord : vec4<i32>) -> f32 {
                  let b = getPreluActivationWeightsByOutputCoords(outCoord);
                  ${c}
                }` : i = `
                  fn activation(a : f32, outCoord : vec4<i32>) -> f32 {
                    ${c}
                  }
                `, o = "value = activation(value, outCoord);";
    }
    let u = this.addBias ? "value = value + getBiasByOutputCoords(outCoord);" : "";
    return `
    ${i}
    fn mm_readA(row : i32, col : i32, globalId : vec3<u32>) -> f32 {
      var batch = i32(globalId.z);
      ${r}
    }

    fn mm_readB(row : i32, col : i32, globalId : vec3<u32>) -> f32 {
      ${a}
    }

    fn mm_write(row : i32, col : i32, valueInput : f32, globalId : vec3<u32>) {
      var batch = i32(globalId.z);
      var value = valueInput;
      let outWidth = ${this.isChannelsLast ? "uniforms.outShape[2]" : "uniforms.outShape[3]"};
      ${t}
      ${u}
      ${o}
      result[getIndexFromCoords4D(outCoord, uniforms.outShape)] = value;
    }
    ${n}
  `;
  }
};
function wre({ x: e, filter: t, convInfo: n, backend: s, bias: r = null, preluActivationWeights: a = null, leakyreluAlpha: i = 0, activation: o = null }) {
  let u = n.dataFormat === "channelsLast", l = !u, c = false, p = u && n.filterHeight === n.inHeight && n.filterWidth === n.inWidth && n.padInfo.type === "VALID", d, h;
  if (p) {
    let g = n.inHeight * n.inWidth * n.inChannels;
    d = We({ inputs: { x: e }, backend: s, attrs: { shape: [1, n.batchSize, g] } }), h = We({ inputs: { x: t }, backend: s, attrs: { shape: [1, g, n.outChannels] } });
  } else
    d = We({ inputs: { x: e }, backend: s, attrs: { shape: u ? [n.batchSize, n.inHeight * n.inWidth, n.inChannels] : [n.batchSize, n.inChannels, n.inHeight * n.inWidth] } }), h = We({ inputs: { x: t }, backend: s, attrs: { shape: [1, n.inChannels, n.outChannels] } });
  let f = Pv({ a: u ? d : h, b: u ? h : d, transposeA: l, transposeB: c, backend: s, bias: r, activation: o, preluActivationWeights: a, leakyreluAlpha: i }), m = We({ inputs: { x: f }, backend: s, attrs: { shape: n.outShape } });
  return s.disposeData(d.dataId), s.disposeData(h.dataId), s.disposeData(f.dataId), m;
}
function P2({ x: e, filter: t, convInfo: n, backend: s, bias: r = null, preluActivationWeights: a = null, leakyreluAlpha: i = 0, activation: o = null }) {
  let u = r != null, l = a != null, c = n.dataFormat === "channelsLast", p;
  if (c && n.filterHeight === n.inHeight && n.filterWidth === n.inWidth && n.padInfo.type === "VALID" || n.filterHeight === 1 && n.filterWidth === 1 && n.dilationHeight === 1 && n.dilationWidth === 1 && n.strideHeight === 1 && n.strideWidth === 1 && (n.padInfo.type === "SAME" || n.padInfo.type === "VALID"))
    return wre({ x: e, filter: t, convInfo: n, backend: s, bias: r, activation: o, preluActivationWeights: a, leakyreluAlpha: i });
  let h = (n.inChannels % 4 === 0 || n.inChannels === 3 && n.padInfo.type === "VALID") && n.outChannels % 4 === 0 && c, f = [n.padInfo.top, n.padInfo.left], m = [{ type: "int32", data: [n.filterHeight, n.filterWidth] }, { type: "int32", data: [...f] }, { type: "int32", data: [n.strideHeight, n.strideWidth] }, { type: "int32", data: [n.dilationHeight, n.dilationWidth] }];
  h ? p = new vre(n, u, o, l) : p = new xre(n, u, o, l);
  let g = n.outHeight * n.outWidth, b = n.outChannels, y = n.filterHeight * n.filterWidth * n.inChannels;
  m.push({ type: "int32", data: [g] }, { type: "int32", data: [b] }, { type: "int32", data: [y] });
  let v = [e, t];
  return u && v.push(r), l && v.push(a), o === "leakyrelu" && (m.push({ type: "float32", data: [i] }), p.uniforms += " alpha : f32,"), s.runWebGPUProgram(p, v, e.dtype, m);
}
function kre(e) {
  let { inputs: t, attrs: n, backend: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dataFormat: u, dilations: l, dimRoundingMode: c } = n, p = C.convertConv2DDataFormat(u), d = C.computeConv2DInfo(r.shape, a.shape, i, l, o, c, false, p);
  return P2({ x: r, filter: a, convInfo: d, backend: s });
}
var Sre = { kernelName: _a, backendName: "webgpu", kernelFunc: kre };
var Ire = class {
  constructor(e) {
    this.variableNames = ["x", "W"], this.uniforms = "filterDims : vec2<i32>, pads : vec2<i32>, stride : vec2<i32>, outBackprop : vec4<i32>, dimAOuter : i32, dimBOuter : i32, dimInner : i32,", this.outputShape = e.inShape, w.assert(e.dataFormat === "channelsLast", () => "TODO: NCHW is unimplemented"), this.dispatchLayout = { x: [3], y: [1, 2], z: [0] }, this.workGroupSize = Rv(this.dispatchLayout, this.outputShape), this.elementsPerThread = Dv(this.dispatchLayout, this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, this.elementsPerThread), this.shaderKey = `conv2DDerInputMM_${this.elementsPerThread}`;
  }
  getUserCode() {
    return `
    fn mm_readA(row : i32, col : i32, globalId : vec3<u32>) -> f32 {
      var batch = i32(globalId.z);
      if (row < uniforms.dimAOuter && col < uniforms.dimInner) {
      
    let outRow = row / uniforms.outShape[2];
    let outCol = row % uniforms.outShape[2];

    let WRow = col / (uniforms.filterDims[1] * uniforms.outBackprop[3]);
    let WCol = col / uniforms.outBackprop[3] % uniforms.filterDims[1];
    let xR = f32(outRow - uniforms.pads[0] + WRow) / f32(uniforms.stride[0]);
    let xC = f32(outCol - uniforms.pads[1] + WCol) / f32(uniforms.stride[1]);
    if (xR < 0.0 || xR >= f32(uniforms.outBackprop[1]) || fract(xR) > 0.0) {
      return 0.0;
    }
    if (xC < 0.0 || xC >= f32(uniforms.outBackprop[2]) || fract(xC) > 0.0) {
      return 0.0;
    }
    let coord = vec4<i32>(
        batch,
        i32(xR),
        i32(xC),
        col % uniforms.outBackprop[3]);
    return x[getIndexFromCoords4D(coord, uniforms.xShape)];
    }
    return 0.0;
    }

    fn mm_readB(row : i32, col : i32, globalId : vec3<u32>) -> f32 {
      let coordX = uniforms.filterDims.x - 1 -
          row / (uniforms.filterDims[1] * uniforms.outBackprop[3]);
      let coordY = uniforms.filterDims.y - 1 -
          (row / uniforms.outBackprop[3]) % uniforms.filterDims[1];
      if (row < uniforms.dimInner && col < uniforms.dimBOuter &&
          coordX >= 0 && coordY >= 0) {
        let coord = vec4<i32>(coordX, coordY, col,
            row % uniforms.outBackprop[3]);
        return W[getIndexFromCoords4D(coord, uniforms.wShape)];
      }
      return 0.0;
    }

    fn mm_write(row : i32, col : i32, valueInput : f32, globalId : vec3<u32>) {
      var batch = i32(globalId.z);
      var value = valueInput;
      let outCoord = vec4<i32>(
          batch,
          row / uniforms.outShape[2],
          row % uniforms.outShape[2],
          col);
      result[getIndexFromCoords4D(outCoord, uniforms.outShape)] = value;
    }

    ${Ov(this.elementsPerThread, this.workGroupSize)}
  `;
  }
};
var Cre = class {
  constructor(e) {
    this.variableNames = ["dy", "W"], this.uniforms = "filterDims : vec2<i32>, pads : vec2<i32>, stride : vec2<i32>, outBackprop : vec4<i32>,", this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e.inShape, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.isChannelsLast = e.dataFormat === "channelsLast", this.shaderKey = `conv2DDerInput_${this.isChannelsLast}`;
  }
  getUserCode() {
    let e = this.isChannelsLast ? 1 : 2, t = this.isChannelsLast ? 2 : 3, n = this.isChannelsLast ? 3 : 1;
    return `
    ${Ue()} {
      if(index < uniforms.size) {
        let coords = getCoordsFromIndex(index);
        let batch = coords[0];
        let d1 = coords[${n}];

        let dyCorner = vec2<i32>(coords[${e}]), coords[${t}]) - uniforms.pads;
        let dyRCorner = dyCorner.x;
        let dyCCorner = dyCorner.y;

        // Convolve dy(?, ?, d2) with w(:, :, d1, d2) to compute dx(xR, xC, d1).
        // ? = to be determined. : = across all values in that axis.
        var dotProd = 0.0;
        for (var wR = 0; wR < uniforms.filterDims.x; wR = wR + 1) {
          let dyR = (f32(dyRCorner) + f32(wR)) / f32(uniforms.stride.x);
          let wRPerm = uniforms.filterDims.x - 1 - wR;
          if (dyR < 0.0 || dyR >= f32(uniforms.outBackprop[1]) || fract(dyR) > 0.0 ||
              wRPerm < 0) {
            continue;
          }
          let idyR = dyR;

          for (var wC = 0; wC < uniforms.filterDims.y; wC = wC + 1) {
            let dyC = (f32(dyCCorner) + f32(wC)) / f32(uniforms.stride.y);
            let wCPerm = uniforms.filterDims.y - 1 - wC;
            if (dyC < 0.0 || dyC >= f32(uniforms.outBackprop[2]) ||
                fract(dyC) > 0.0 || wCPerm < 0) {
              continue;
            }
            let idyC = dyC;

            for (var d2 = 0; d2 < uniforms.outBackprop[3]; d2 = d2 + 1) {
              if (${this.isChannelsLast}) {
                let xValue = getDy(batch, idyR, idyC, d2);
                let wValue = getW(wRPerm, wCPerm, d1, d2);
                dotProd = dotProd + xValue * wValue;
              } else {
                let xValue = getDy(batch, d2, idyR, idyC);
                let wValue = getW(wRPerm, wCPerm, d1, d2);
                dotProd = dotProd + xValue * wValue;
              }

            }
          }
        }
        setOutputAtIndex(index, dotProd);
      }
    }
  `;
  }
};
function Nre(e) {
  let { inputs: t, backend: n, attrs: s } = e, { dy: r, filter: a } = t, { inputShape: i, strides: o, pad: u, dataFormat: l, dimRoundingMode: c } = s, p = C.convertConv2DDataFormat(l), d = C.computeConv2DInfo(i, a.shape, o, 1, u, c, false, p), h = [{ type: "int32", data: [d.filterHeight, d.filterWidth] }, { type: "int32", data: [d.filterHeight - 1 - d.padInfo.top, d.filterWidth - 1 - d.padInfo.left] }, { type: "int32", data: [d.strideHeight, d.strideWidth] }, { type: "int32", data: [d.batchSize, d.outHeight, d.outWidth, d.outChannels] }], f;
  if (K().getBool("WEBGPU_USE_NAIVE_CONV2D_TRANSPOSE"))
    f = new Cre(d);
  else {
    f = new Ire(d);
    let m = d.inShape[1] * d.inShape[2], g = d.inShape[3], b = d.filterHeight * d.filterWidth * d.outChannels;
    h.push({ type: "uint32", data: [m] }, { type: "uint32", data: [g] }, { type: "uint32", data: [b] });
  }
  return n.runWebGPUProgram(f, [r, a], "float32", h);
}
var Tre = { kernelName: Aa, backendName: "webgpu", kernelFunc: Nre };
var $re = Kt({ opType: 2 });
var _re = { kernelName: Ea, backendName: "webgpu", kernelFunc: $re };
var Are = Kt({ opType: 3 });
var Ere = { kernelName: Ra, backendName: "webgpu", kernelFunc: Are };
var Rre = class {
  constructor(e, t, n, s) {
    this.variableNames = ["Image", "Boxes", "BoxInd"], this.uniforms = "extrapolationValue : f32,", this.workGroupSize = [64, 1, 1], this.size = true;
    let [r] = t;
    this.outputShape = [r, n[0], n[1], e], this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.methodId = s === "bilinear" ? 1 : 0, this.cropHeightBiggerThan1 = this.outputShape[1] > 1, this.cropWidthBiggerThan1 = this.outputShape[2] > 1, this.shaderKey = `cropAndResize_${this.methodId}_${this.cropHeightBiggerThan1}_${this.cropWidthBiggerThan1}`;
  }
  getUserCode() {
    let [e, t] = ["f32(uniforms.imageShape[1] - 1)", "f32(uniforms.imageShape[2] - 1)"], [n, s, r] = this.cropHeightBiggerThan1 ? [`(${e} / f32(uniforms.outShape[1] - 1))`, "(y2-y1) * height_ratio", `y1*${e} + f32(y)*(height_scale)`] : ["0.0", "0.0", `0.5 * (y1+y2) * ${e}`], [a, i, o] = this.cropWidthBiggerThan1 ? [`(${t} / f32(uniforms.outShape[2] - 1))`, "(x2-x1) * width_ratio", `x1*${t} + f32(x)*(width_scale)`] : ["0.0", "0.0", `0.5 * (x1+x2) * ${t}`];
    return `
      ${Ue()}
      if (index < uniforms.size) {
        let coords = getCoordsFromIndex(index);
        let height_ratio = f32(${n});
        let width_ratio = f32(${a});
        let b = coords[0];
        let y = coords[1];
        let x = coords[2];
        let d = coords[3];
        // get box vals
        let y1 = getBoxes(b, 0);
        let x1 = getBoxes(b, 1);
        let y2 = getBoxes(b, 2);
        let x2 = getBoxes(b, 3);
        // get image in batch index
        let bInd = i32(round(getBoxInd(b)));
        if(bInd < 0 || bInd >= uniforms.outShape[0]) {
          return;
        }
        let height_scale = ${s};
        let width_scale = ${i};
        let in_y = ${r};
        if( in_y < 0.0 || in_y > ${e} ) {
          setOutputAtIndex(index, uniforms.extrapolationValue);
          return;
        }
        let in_x = ${o};
        if( in_x < 0.0 || in_x > ${t} ) {
          setOutputAtIndex(index, uniforms.extrapolationValue);
          return;
        }
        let sourceFracIndexCR = vec2<f32>(in_x,in_y);
        if(${this.methodId} == 1) {
          // Compute the four integer indices.
          let sourceFloorCR = vec2<i32>(sourceFracIndexCR);
          let sourceCeilCR = vec2<i32>(ceil(sourceFracIndexCR));
          let topLeft = getImage(bInd, sourceFloorCR.y, sourceFloorCR.x, d);
          let bottomLeft = getImage(bInd, sourceCeilCR.y, sourceFloorCR.x, d);
          let topRight = getImage(bInd, sourceFloorCR.y, sourceCeilCR.x, d);
          let bottomRight = getImage(bInd, sourceCeilCR.y, sourceCeilCR.x, d);
          let fracCR = sourceFracIndexCR - vec2<f32>(sourceFloorCR);
          let top = topLeft + (topRight - topLeft) * fracCR.x;
          let bottom = bottomLeft + (bottomRight - bottomLeft) * fracCR.x;
          let newValue = top + (bottom - top) * fracCR.y;
          setOutputAtIndex(index, newValue);
        } else {
          // Compute the coordinators of nearest neighbor point.
          let sourceNearestCR = vec2<i32>(floor(
            sourceFracIndexCR + vec2<f32>(0.5,0.5)));
          let newValue = getImage(
            bInd, sourceNearestCR.y, sourceNearestCR.x, d);
          setOutputAtIndex(index, newValue);
        }
      }
    }
    `;
  }
};
var Dre = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { image: r, boxes: a, boxInd: i } = t, { cropSize: o, method: u, extrapolationValue: l } = s, c = new Rre(r.shape[3], a.shape, o, u), p = [{ type: "float32", data: [l] }];
  return n.runWebGPUProgram(c, [r, a, i], "float32", p);
};
var Fre = { kernelName: go, backendName: "webgpu", kernelFunc: Dre };
var Mw = class {
  constructor(e, t, n, s) {
    this.variableNames = ["x"], this.uniforms = "index : f32,", this.size = true;
    let r = 128;
    this.workGroupSize = [r, 1, 1], this.outputShape = t, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.exclusive = n, this.reverse = s, this.op = e, this.shaderKey = `cum_${this.op}_${this.exclusive}_${this.reverse}`;
  }
  getUserCode() {
    let e = this.outputShape.length, t = this.op === "*" ? "1.0" : "0.0", n = this.exclusive ? t : `getX(${Lw(e, "coords", this.op)})`, s = this.outputShape[this.outputShape.length - 1], r = "", a = "";
    return this.exclusive ? (r = this.reverse ? `end != ${s - 1}` : "end != 0", a = this.reverse ? "end + 1" : "end - 1") : (r = this.reverse ? `end + pow2 < ${s}` : "end >= pow2", a = this.reverse ? "end + pow2" : "end - pow2"), `
      ${Ue()}
       if (index < uniforms.size) {
         var coords = getCoordsFromIndex(index);

         let end = ${Bw(e, "coords", this.op)};
         var val = ${n};
         let pow2 = i32(pow(2.0, uniforms.index));
         if (${r}) {
           let idx = ${a};
           ${Bw(e, "coords", this.op)} = idx;
           val ${this.op}= getX(${Lw(e, "coords", this.op)});
         }
         setOutputAtIndex(index, val);
       }
      }
    `;
  }
};
function Lw(e, t, n) {
  if (e === 1)
    return `${t}`;
  if (e === 2)
    return `${t}.x, ${t}.y`;
  if (e === 3)
    return `${t}.x, ${t}.y, ${t}.z`;
  if (e === 4)
    return `${t}.x, ${t}.y, ${t}.z, ${t}.w`;
  throw Error(`Cumulative ${n} for rank ${e} is not yet supported`);
}
function Bw(e, t, n) {
  if (e === 1)
    return `${t}`;
  if (e === 2)
    return `${t}.y`;
  if (e === 3)
    return `${t}.z`;
  if (e === 4)
    return `${t}.w`;
  throw Error(`Cumulative ${n} for rank ${e} is not yet supported`);
}
function z2(e, t, n, s, r, a) {
  let i = t.shape.length, o = C.getAxesPermutation([s], i), u = t;
  o != null && (u = Ks({ inputs: { x: t }, backend: n, attrs: { perm: o } }));
  let l = C.getInnerMostAxes(1, i)[0];
  if (l !== i - 1)
    throw new Error(`WebGPU cumprod shader expects an inner-most axis=${t.shape.length - 1} but got axis=${s}`);
  let c = u.shape[l], p = Wn({ inputs: { x: u }, backend: n });
  for (let d = 0; d <= Math.ceil(Math.log2(c)) - 1; d++) {
    let h = new Mw(e, u.shape, false, a), f = p, m = [{ type: "float32", data: [d] }];
    p = n.runWebGPUProgram(h, [p], p.dtype, m), n.disposeData(f.dataId);
  }
  if (r) {
    let d = new Mw(e, u.shape, r, a), h = p, f = [{ type: "float32", data: [0] }];
    p = n.runWebGPUProgram(d, [p], p.dtype, f), n.disposeData(h.dataId);
  }
  if (o != null) {
    let d = C.getUndoAxesPermutation(o), h = Ks({ inputs: { x: p }, backend: n, attrs: { perm: d } });
    return n.disposeData(p.dataId), n.disposeData(u.dataId), h;
  }
  return p;
}
function Ore(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, exclusive: i, reverse: o } = s;
  return z2("*", r, n, a, i, o);
}
var Pre = { kernelName: mo, backendName: "webgpu", kernelFunc: Ore };
function zre(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, exclusive: i, reverse: o } = s;
  return z2("+", r, n, a, i, o);
}
var Mre = { kernelName: Da, backendName: "webgpu", kernelFunc: zre };
var Lre = class {
  constructor(e, t) {
    this.variableNames = ["x"], this.workGroupSize = [64, 1, 1], this.size = true, this.uniforms = "blockSize : i32,", this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = `depthToSpace_${t}`, this.dataFormat = t;
  }
  getUserCode() {
    return `
      ${Ue()}
        if (index < uniforms.size) {
          let coords = getCoordsFromIndex(index);
          let b = coords[0];
          let h = ${this.getHeightCoordString()};
          let w = ${this.getWidthCoordString()};
          let d = ${this.getDepthCoordString()};

          let in_h = h / uniforms.blockSize;
          let offset_h = h % uniforms.blockSize;
          let in_w = w / uniforms.blockSize;
          let offset_w = w % uniforms.blockSize;
          let offset_d = (offset_h * uniforms.blockSize + offset_w) *
            ${this.getOutputDepthSize()};
          let in_d = d + offset_d;

          let rlt = ${this.getInputSamplingString()};
          setOutputAtIndex(index, rlt);
        }
      }`;
  }
  getHeightCoordString() {
    return this.dataFormat === "NHWC" ? "coords[1]" : "coords[2]";
  }
  getWidthCoordString() {
    return this.dataFormat === "NHWC" ? "coords[2]" : "coords[3]";
  }
  getDepthCoordString() {
    return this.dataFormat === "NHWC" ? "coords[3]" : "coords[1]";
  }
  getOutputDepthSize() {
    return this.dataFormat === "NHWC" ? "uniforms.outShape[3]" : "uniforms.outShape[1]";
  }
  getInputSamplingString() {
    return this.dataFormat === "NHWC" ? "getX(b, in_h, in_w, in_d)" : "getX(b, in_d, in_h, in_w)";
  }
};
function Bre(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockSize: a, dataFormat: i } = s, o = r.shape[0], u = i === "NHWC" ? r.shape[1] : r.shape[2], l = i === "NHWC" ? r.shape[2] : r.shape[3], c = i === "NHWC" ? r.shape[3] : r.shape[1], p = u * a, d = l * a, h = c / (a * a), f = i === "NHWC" ? [o, p, d, h] : [o, h, p, d], m = [{ type: "int32", data: [a] }], g = new Lre(f, i);
  return n.runWebGPUProgram(g, [r], r.dtype, m);
}
var Vre = { kernelName: bo, backendName: "webgpu", kernelFunc: Bre };
var M2 = class {
  constructor(e, t = false, n = null, s = false) {
    this.variableNames = ["x", "W"], this.uniforms = "pad : vec2<i32>, stride : vec2<i32>, dilation : vec2<i32>, inDims : vec2<i32>,", this.workGroupSize = [4, 4, 4], this.isVec4 = true, this.outputShape = e.outShape, this.dispatchLayout = { x: [0, 1], y: [2], z: [3] }, this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [1, 4, 4]), w.assert(e.dataFormat === "channelsLast", () => "TODO: NCHW is unimplemented"), t && this.variableNames.push("bias"), s && this.variableNames.push("preluActivationWeights"), this.convInfo = e, this.addBias = t, this.activation = n, this.hasPreluActivation = s, this.shaderKey = `depthwise3x3_${n}`;
  }
  getUserCode() {
    let e = "", t = "";
    if (this.activation) {
      let r = Fr(this.activation, this.isVec4);
      this.hasPreluActivation ? e = `fn activation(a : vec4<f32>, outCoord : vec4<i32>) -> vec4<f32> {
          let b = getPreluActivationWeightsByOutputCoords(outCoord);
          ${r}
        }` : e = `
        fn activation(a : vec4<f32>, outCoord : vec4<i32>) -> vec4<f32> {
            ${r}
          }
        `, t = "dotProd[i] = activation(dotProd[i], coords);";
    }
    let n = this.addBias ? "dotProd[i] = dotProd[i] + getBiasByOutputCoords(coords);" : "";
    return `
      ${e}

      ${Ev()}
      fn main(@builtin(global_invocation_id) globalId: vec3<u32>) {
        let batch = 0;
        let r = i32(globalId.x);
        let c = i32(globalId.y) * 4;
        let d2 = i32(globalId.z) * 4;
        let xRCCorner = vec2<i32>(r, c) * uniforms.stride - uniforms.pad;
        let d1 = d2;
        let q = 0;

        let xRCorner = xRCCorner.x;
        let xCCorner = xRCCorner.y;

        var wVals : array<vec4<f32>, 9>;
        wVals[0] = getW(0, 0, d1, q);
        wVals[1] = getW(0, 1, d1, q);
        wVals[2] = getW(0, 2, d1, q);
        wVals[3] = getW(1, 0, d1, q);
        wVals[4] = getW(1, 1, d1, q);
        wVals[5] = getW(1, 2, d1, q);
        wVals[6] = getW(2, 0, d1, q);
        wVals[7] = getW(2, 1, d1, q);
        wVals[8] = getW(2, 2, d1, q);

        var xVals : array<array<vec4<f32>, 6>, 3>;
        for (var wR = 0; wR < 3; wR = wR + 1) {
          let xR = xRCorner + wR * uniforms.dilation[0];
          for (var wC = 0; wC < 6; wC = wC + 1) {
            let xC = xCCorner + wC * uniforms.dilation[1];
            if (xR < 0 || xR >= uniforms.inDims[0] || xC < 0 || xC >= uniforms.inDims[1]) {
              xVals[wR][wC] = vec4<f32>(0.0);
            } else {
              xVals[wR][wC] = getX(batch, xR, xC, d1);
            }
          }
        }

        var dotProd : array<vec4<f32>, 4>;
        dotProd[0] = vec4<f32>(0.0);
        dotProd[1] = vec4<f32>(0.0);
        dotProd[2] = vec4<f32>(0.0);
        dotProd[3] = vec4<f32>(0.0);

        for (var wR = 0; wR < 3; wR = wR + 1) {
          for (var wC = 0; wC < 3; wC = wC + 1) {
            let indexW = wR * 3 + wC;
            dotProd[0] = dotProd[0] + xVals[wR][0 + wC] * wVals[indexW];
            dotProd[1] = dotProd[1] + xVals[wR][1 + wC] * wVals[indexW];
            dotProd[2] = dotProd[2] + xVals[wR][2 + wC] * wVals[indexW];
            dotProd[3] = dotProd[3] + xVals[wR][3 + wC] * wVals[indexW];
          }
        }

        for (var i = 0; i < 4; i = i + 1) {
          let coords = vec4<i32>(batch, r, c + i, d2);
          if (coordsInBounds4D(coords, uniforms.outShape)) {
            ${n}
            ${t}
            setOutputAtCoords(coords[0], coords[1], coords[2], coords[3], dotProd[i]);
          }
        }
      }
    `;
  }
};
var L2 = class {
  constructor(e, t = false, n = null, s = false) {
    this.variableNames = ["x", "W"], this.uniforms = `pad : vec2<i32>, stride : vec2<i32>, dilation : vec2<i32>,
      inDims : vec2<i32>, filterHeight : i32, filterWidth : i32,
      channelMul : i32,`, this.workGroupSize = [256, 1, 1], this.outputShape = e.outShape, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), w.assert(e.dataFormat === "channelsLast", () => "TODO: NCHW is unimplemented"), t && this.variableNames.push("bias"), s && this.variableNames.push("preluActivationWeights"), this.convInfo = e, this.addBias = t, this.activation = n, this.hasPreluActivation = s, this.shaderKey = `depthwise_${this.activation}`;
  }
  getUserCode() {
    let e = "", t = "";
    if (this.activation) {
      let r = Fr(this.activation, false);
      this.hasPreluActivation ? e = `fn activation(a : f32, outCoord : vec4<i32>) -> f32 {
          let b = getPreluActivationWeightsByOutputCoords(outCoord);
          ${r}
        }` : e = `
          fn activation(a : f32, outCoord : vec4<i32>) -> f32 {
            ${r}
          }
        `, t = "dotProd = activation(dotProd, coords);";
    }
    let n = this.addBias ? "dotProd = dotProd + getBiasByOutputCoords(coords);" : "";
    return `
      ${e}

      fn writeResult(batch : i32, row : i32, col : i32, chan : i32,
          value : f32) {
        let coord = vec4<i32>(batch, row, col, chan);
        if (coordsInBounds4D(coord, uniforms.outShape)) {
          setOutputAtCoords(batch, row, col, chan, value);
        }
      }

      ${Ii()}
        let coords = getOutputCoords();
        let batch = coords[0];
        let xRCCorner = vec2<i32>(coords.yz) * uniforms.stride - uniforms.pad;
        let d2 = coords[3];
        let d1 = d2 / uniforms.channelMul;
        let q = d2 - d1 * uniforms.channelMul;

        let inputRowStart = xRCCorner.x;
        let inputColStart = xRCCorner.y;
        let inputRowEnd = inputRowStart + uniforms.filterHeight *
            uniforms.dilation[0];
        let inputColEnd = inputColStart + uniforms.filterWidth *
            uniforms.dilation[1];

        // Convolve x(?, ?, d1) with w(:, :, d1, q) to get y(yR, yC, d2).
        // ? = to be determined. : = across all values in that axis.
        var dotProd = 0.0;

        // Extract if checking out of for loop for performance.
        if (inputRowStart >= 0 && inputColStart >= 0 &&
          inputRowEnd < uniforms.inDims[0] &&
              inputColEnd < uniforms.inDims[1]) {
            // Here using a constant value |this.convInfo.filterHeight| instead
            // of uniform value is in order to loop unrolling.
            for (var wR = 0; wR < uniforms.filterHeight; wR = wR + 1) {
              let xR = inputRowStart + wR * uniforms.dilation[0];

              for (var wC = 0; wC < uniforms.filterWidth; wC = wC + 1) {
                let xC = inputColStart + wC * uniforms.dilation[1];

                let xVal = getX(batch, xR, xC, d1);
                let wVal = getW(wR, wC, d1, q);
                dotProd = dotProd + xVal * wVal;
              }
            }
          } else {
            for (var wR = 0; wR < uniforms.filterHeight; wR = wR + 1) {
              let xR = inputRowStart + wR * uniforms.dilation[0];

              if (xR < 0 || xR >= uniforms.inDims[0]) {
                continue;
              }

              for (var wC = 0; wC < uniforms.filterWidth; wC = wC + 1) {
                let xC = inputColStart + wC * uniforms.dilation[1];

                if (xC < 0 || xC >= uniforms.inDims[1]) {
                  continue;
                }

                let xVal = getX(batch, xR, xC, d1);
                let wVal = getW(wR, wC, d1, q);
                dotProd = dotProd + xVal * wVal;
              }
            }
          }

        ${n}
        ${t}
        writeResult(batch, coords[1], coords[2], d2, dotProd);
      }
    `;
  }
};
function Wre(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a } = t, { strides: i, pad: o, dilations: u, dimRoundingMode: l } = s, c = u;
  c == null && (c = [1, 1]);
  let p = C.computeConv2DInfo(r.shape, a.shape, i, c, o, l, true), d = [{ type: "int32", data: [p.padInfo.top, p.padInfo.left] }, { type: "int32", data: [p.strideHeight, p.strideWidth] }, { type: "int32", data: [p.dilationHeight, p.dilationWidth] }, { type: "int32", data: [p.inHeight, p.inWidth] }], h;
  return p.batchSize === 1 && p.inHeight === p.outHeight && p.inWidth === p.outWidth && p.strideHeight === 1 && p.strideWidth === 1 && p.filterHeight === p.filterWidth && p.inChannels === p.outChannels && p.dilationHeight === 1 && p.dilationWidth === 1 && p.filterHeight === 3 && p.inChannels % 4 === 0 ? h = new M2(p) : (h = new L2(p), d.push({ type: "int32", data: [p.filterHeight] }, { type: "int32", data: [p.filterWidth] }, { type: "int32", data: [p.outChannels / p.inChannels] })), n.runWebGPUProgram(h, [r, a], r.dtype, d);
}
var Ure = { kernelName: Fa, backendName: "webgpu", kernelFunc: Wre };
var B2 = mn({ opSnippet: 0, cpuKernelImpl: xse, supportsComplex: true });
var Gre = { kernelName: Za, backendName: "webgpu", kernelFunc: B2 };
var Hre = class {
  constructor(e, t) {
    this.workGroupSize = [64, 1, 1], this.variableNames = ["x"], this.uniforms = "reduceSize : i32,", this.size = true, this.inputShape = [e.batchSize, e.inSize];
    let [n] = C.computeOutAndReduceShapes(this.inputShape, [1]);
    this.outputShape = n.length === 0 ? [1] : n, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, [1, 1, 1]), this.reduceType = t, this.shaderKey = `reduce_${t}`;
  }
  getUserCode() {
    let e = "", t = "0.0";
    this.reduceType === "min" || this.reduceType === "max" ? (e = `
         if (isnan(candidate)) {
          bestValue = uniforms.NAN;
         } else if (!isnan(bestValue) && candidate ${this.reduceType === "min" ? "<" : ">"} bestValue)
           {  bestValue = candidate; }`, t = "f32(x[offset])") : this.reduceType === "sum" || this.reduceType === "mean" ? e = " bestValue = bestValue + candidate; " : this.reduceType === "prod" && (e = " bestValue = bestValue * candidate; ", t = "1.0");
    let n = this.reduceType === "mean" ? "setOutputAtIndex(outputIndex, bestValue / f32(uniforms.reduceSize));" : "setOutputAtIndex(outputIndex, bestValue);";
    return `
       fn DIV_CEIL(a : u32, b : u32) -> u32 {
        return ((a - 1u) / b + 1u);
       }

       ${`
         var<workgroup> xBestValues : array<f32, ${this.workGroupSize[0]}>;
       `}
       fn getOffset(outputIndex : i32) -> i32 {
         let outputCoords = getCoordsFromIndex(outputIndex);
         let offset = ${this.outputShape.length === 1 ? "outputCoords" : "outputCoords[0]"} * uniforms.reduceSize;
          return offset;
       }
       ${Ue()}
         let outputIndex = index / i32(workGroupSizeX);
         let offset = getOffset(outputIndex);
         var bestValue = ${t};
         let Length = uniforms.reduceSize;
         let WorkPerThread = DIV_CEIL(u32(Length), workGroupSizeX);
         for (var k = i32(localId.x); k < Length && outputIndex < uniforms.size;
             k = k + i32(workGroupSizeX)) {
           let candidate = f32(x[offset + k]);
           ${e}
         }
         xBestValues[localId.x] = bestValue;
         workgroupBarrier();

         var reduceSize = min(u32(Length), workGroupSizeX);
         for (var currentSize = reduceSize / 2u; reduceSize > 1u;
             currentSize = reduceSize / 2u) {
           let interval = DIV_CEIL(reduceSize, 2u);
           if (localId.x < currentSize) {
            let candidate = xBestValues[localId.x + interval];
            ${e}
            xBestValues[localId.x] = bestValue;
           }
           reduceSize = interval;
           workgroupBarrier();
         }

         if (localId.x == 0u && outputIndex < uniforms.size) {
          ${n}
        }
       }
     `;
  }
};
function ic(e, t, n, s, r) {
  let a = e.shape.length, i = [], o = w.parseAxisParam(t, e.shape), u = o, l = C.getAxesPermutation(u, a), c = e;
  l != null && (c = Ks({ inputs: { x: e }, attrs: { perm: l }, backend: r }), u = C.getInnerMostAxes(u.length, a), i.push(c)), C.assertAxesAreInnerMostDims(s, u, a);
  let [p, d] = C.computeOutAndReduceShapes(c.shape, u), h = p;
  n && (h = C.expandShapeToKeepDim(p, o));
  let f;
  if ((s === "max" || s === "prod") && r.shouldExecuteOnCPU([c])) {
    let m = r.tensorMap.get(c.dataId).values;
    switch (s) {
      case "max":
        let g = bse(m, w.sizeFromShape(d), h, e.dtype);
        f = r.makeTensorInfo(h, e.dtype, g);
        break;
      case "prod":
        let { outVals: b, outShape: y, outDtype: v } = Sse(c.shape, c.dtype, m, u);
        f = r.makeTensorInfo(y, v, b);
        break;
      default:
        throw new Error(`${s} CPU implementation is not yet supported.`);
    }
  } else {
    let m = w.sizeFromShape(d), b = w.sizeFromShape(c.shape) / m, y = { windowSize: m, inSize: m, batchSize: b, outSize: 1 }, v = s === "mean" ? "float32" : bp(e.dtype), x = [{ type: "int32", data: [m] }], k = new Hre(y, s), I = r.runWebGPUProgram(k, [c], v, x);
    i.push(I), f = We({ inputs: { x: I }, attrs: { shape: h }, backend: r });
  }
  return i.forEach((m) => r.disposeData(m.dataId)), f;
}
function zv(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  return ic(r, a, i, "sum", n);
}
var qre = { kernelName: ci, backendName: "webgpu", kernelFunc: zv };
function jre(e) {
  let { inputs: t, backend: n, attrs: s } = e, { equation: r } = s, a = t, { allDims: i, summedDims: o, idDims: u } = C.decodeEinsumEquation(r, a.length);
  C.checkEinsumDimSizes(i.length, u, a);
  let { path: l, steps: c } = C.getEinsumComputePath(o, u), p = c.length, d = null, h = i.length, f = [];
  for (let m = 0; m < p; ++m) {
    for (let g of c[m]) {
      let { permutationIndices: b, expandDims: y } = C.getEinsumPermutation(h, u[g]), v;
      C.isIdentityPermutation(b) ? v = a[g] : (v = Ks({ inputs: { x: a[g] }, backend: n, attrs: { perm: b } }), f.push(v));
      let x = v.shape.slice();
      for (let k = 0; k < y.length; ++k)
        x.splice(y[k], 0, 1);
      w.arraysEqual(v.shape, x) || (v = We({ inputs: { x: v }, backend: n, attrs: { shape: x } }), f.push(v)), d === null ? d = v : (d = B2({ inputs: { a: v, b: d }, backend: n }), f.push(d));
    }
    m < p - 1 && (l[m] >= 0 && (d = zv({ inputs: { x: d }, backend: n, attrs: { axis: l[m] - (i.length - h), keepDims: false } }), f.push(d)), h--);
  }
  for (let m of f)
    m !== d && n.disposeData(m.dataId);
  return d;
}
var Kre = { kernelName: rp, backendName: "webgpu", kernelFunc: jre };
var Xre = Kt({ opType: 4 });
var Yre = { kernelName: Pa, backendName: "webgpu", kernelFunc: Xre };
var Qre = mn({ opSnippet: 4, dtype: "bool", cpuKernelImpl: ise });
var Zre = { kernelName: yo, backendName: "webgpu", kernelFunc: Qre };
var V2 = Kt({ opType: 5, cpuKernelImpl: ose, dtype: "float32" });
var Jre = { kernelName: za, backendName: "webgpu", kernelFunc: V2 };
function sg(e) {
  let { inputs: t, attrs: n, backend: s } = e, { dim: r } = n, { input: a } = t, i = a.shape.length, o = a.shape.slice(), u = r;
  return r < 0 && (w.assert(-(i + 1) <= r, () => `Axis must be in the interval [${-(i + 1)}, ${i}]`), u = i + r + 1), o.splice(u, 0, 1), We({ inputs: { x: a }, backend: s, attrs: { shape: o } });
}
var eae = { kernelName: vo, backendName: "webgpu", kernelFunc: sg };
var tae = Kt({ opType: 6, cpuKernelImpl: use });
var nae = { kernelName: xo, backendName: "webgpu", kernelFunc: tae };
var sae = class {
  constructor(e) {
    this.variableNames = [], this.outputShape = [], this.uniforms = "value : f32,", this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = "fill";
  }
  getUserCode() {
    return `
    ${Ue()}
      if (index < uniforms.size) {
        setOutputAtIndex(index, uniforms.value);
      }
    }
  `;
  }
};
function mu(e) {
  let { backend: t, attrs: n } = e, { shape: s, value: r } = n, { dtype: a } = n;
  if (a = a || w.inferDtype(r), a === "string") {
    let i = w.getArrayFromDType(a, w.sizeFromShape(s));
    return i.fill(r), t.makeTensorInfo(s, a, i);
  } else {
    let i = new sae(s), o = [{ type: "float32", data: [r] }];
    return t.runWebGPUProgram(i, [], a, o);
  }
}
var rae = { kernelName: yl, backendName: "webgpu", kernelFunc: mu };
var aae = class {
  constructor(e) {
    this.outputShape = [], this.variableNames = ["x"], this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = "flipLeftRight";
  }
  getUserCode() {
    return `
      ${Ue()}
        if (index < uniforms.size) {
          let coords = getCoordsFromIndex(index);
          let coordX = uniforms.xShape[2] - coords[2] - 1;
          let outputValue = getX(coords[0], coords[1], coordX, coords[3]);
          setOutputAtIndex(index, outputValue);
        }
      }
    `;
  }
};
var iae = { kernelName: wo, backendName: "webgpu", kernelFunc: ({ inputs: e, backend: t }) => {
  let { image: n } = e, s = t, r = new aae(n.shape);
  return s.runWebGPUProgram(r, [n], n.dtype);
} };
var oae = Kt({ opType: 7, cpuKernelImpl: lse });
var uae = { kernelName: Ma, backendName: "webgpu", kernelFunc: oae };
var lae = mn({ opSnippet: 12, dtype: "int32" });
var cae = { kernelName: La, backendName: "webgpu", kernelFunc: lae };
var dae = class {
  constructor(e, t = false) {
    this.outputShape = [0], this.variableNames = [], this.workGroupSize = [256, 1, 1], this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.useImport = t, this.shaderKey = `fromPixels_${this.useImport}`;
  }
  getUserCode() {
    let e = this.useImport ? "textureLoad(src, vec2<i32>(coords.yx));" : "textureLoad(src, vec2<i32>(coords.yx), 0)";
    return `
      @binding(1) @group(0) var src: ${this.useImport ? "texture_external" : "texture_2d<f32>"};

      ${Ue()}
        let flatIndexBase = index * uniforms.numChannels;
        for (var i = 0; i < uniforms.numChannels; i = i + 1) {
          let flatIndex = flatIndexBase + i;
          if (flatIndex < uniforms.size) {
            let coords = getCoordsFromIndex(flatIndexBase);
            let values = ${e};
            result[flatIndex] = i32(floor(255.0 * values[i]));
          }
        }
      }
  `;
  }
};
var pae = { kernelName: yd, backendName: "webgpu", kernelFunc: hae };
var Ui;
function hae(e) {
  let { inputs: t, backend: n, attrs: s } = e, { pixels: r } = t, { numChannels: a } = s;
  if (r == null)
    throw new Error("pixels passed to tf.browser.fromPixels() can not be null");
  let i = typeof HTMLVideoElement != "undefined" && r instanceof HTMLVideoElement, o = typeof HTMLImageElement != "undefined" && r instanceof HTMLImageElement, u = typeof HTMLCanvasElement != "undefined" && r instanceof HTMLCanvasElement || typeof OffscreenCanvas != "undefined" && r instanceof OffscreenCanvas, l = typeof ImageBitmap != "undefined" && r instanceof ImageBitmap, [c, p] = i ? [r.videoWidth, r.videoHeight] : [r.width, r.height], d = [p, c, a];
  if (K().getBool("WEBGPU_USE_IMPORT") && i)
    return Vw({ externalImage: r, backend: n, attrs: s, outShape: d, useImport: true });
  if ((i || o) && (Ui == null && (Ui = document.createElement("canvas").getContext("2d")), Ui.canvas.width = c, Ui.canvas.height = p, Ui.drawImage(r, 0, 0, c, p), r = Ui.canvas), l || u || i || o)
    return Vw({ externalImage: r, backend: n, attrs: s, outShape: d, useImport: false });
  let h = r.data, f = h;
  if (a != null && a !== 4) {
    f = new Uint8Array(r.width * r.height * a);
    let b = h.length, y = 0;
    for (let v = 0; v < b; v++)
      v % 4 < a && (f[y++] = h[v]);
  }
  let m = n.makeTensorInfo(d, "int32"), g = n.tensorMap.get(m.dataId);
  return g.values = new Int32Array(f), n.maybeReleaseBuffer(m.dataId), n.uploadToGPU(m.dataId), m;
}
function Vw(e) {
  let { externalImage: t, backend: n, attrs: s, outShape: r, useImport: a } = e, { numChannels: i } = s, o = w.sizeFromShape(r), u = w.computeStrides(r), l = new dae(r, a), c = [{ type: "uint32", data: [o] }, { type: "uint32", data: [i] }, { type: "uint32", data: [...u] }, { type: "uint32", data: [...l.dispatch] }];
  return n.runFromPixelsProgram(l, r, c, a, t);
}
var fae = class {
  constructor(e, t, n, s, r) {
    this.uniforms = "varianceEpsilon : f32,", this.workGroupSize = [128, 1, 1], this.size = true, this.variableNames = ["x", "mean", "variance"], C.assertAndGetBroadcastShape(e, t), C.assertAndGetBroadcastShape(e, n), this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), s != null && (C.assertAndGetBroadcastShape(e, s), this.variableNames.push("offset")), r != null && (C.assertAndGetBroadcastShape(e, r), this.variableNames.push("scale")), this.offsetShape = s, this.scaleShape = r, this.shaderKey = "batchNorm";
  }
  getUserCode() {
    let e = "0.0";
    this.offsetShape != null && (e = "getOffsetByOutputIndex(index)");
    let t = "1.0";
    return this.scaleShape != null && (t = "getScaleByOutputIndex(index)"), `
      ${Ue()}
        if (index < uniforms.size)
        {
          let xValue = getXByOutputIndex(index);
          let meanValue = getMeanByOutputIndex(index);
          let varianValue = getVarianceByOutputIndex(index);
          let offsetValue = ${e};
          let scaleValue = ${t};
          let inv = scaleValue * inverseSqrt(varianValue + f32(uniforms.varianceEpsilon));
          setOutputAtIndex(index,dot(vec3<f32>(xValue, -meanValue, offsetValue), vec3<f32>(inv, inv, 1.0)));
        }
      }
  `;
  }
};
var mae = { kernelName: Ba, backendName: "webgpu", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { x: s, scale: r, offset: a, mean: i, variance: o } = e, { varianceEpsilon: u } = t, l = n, c = [s, i, o], p = null;
  a != null && (p = a.shape, c.push(a));
  let d = null;
  r != null && (d = r.shape, c.push(r));
  let h = new fae(s.shape, i.shape, o.shape, p, d), f = [{ type: "float32", data: [u] }];
  return l.runWebGPUProgram(h, c, s.dtype, f);
} };
function gae(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a, bias: i, preluActivationWeights: o } = t, { strides: u, pad: l, dataFormat: c, dilations: p, dimRoundingMode: d, activation: h, leakyreluAlpha: f } = s;
  if (c !== "NHWC")
    throw new Error(`WebGPU backend FusedConv2D does not support dataFormat:'${c}'. Please use 'NHWC'.`);
  let m = C.convertConv2DDataFormat(c), g = C.computeConv2DInfo(r.shape, a.shape, u, p, l, d, false, m);
  return P2({ x: r, filter: a, convInfo: g, backend: n, bias: i, preluActivationWeights: o, leakyreluAlpha: f, activation: h });
}
var bae = { kernelName: ia, backendName: "webgpu", kernelFunc: gae };
function yae(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, filter: a, bias: i, preluActivationWeights: o } = t, { strides: u, pad: l, dilations: c, dimRoundingMode: p, activation: d, leakyreluAlpha: h } = s, f = c;
  f == null && (f = [1, 1]), w.assert(C.eitherStridesOrDilationsAreOne(u, f), () => `Error in depthwiseConv2d: Either strides or dilations must be 1. Got strides ${u} and dilations '${f}'`);
  let m = C.computeConv2DInfo(r.shape, a.shape, u, f, l, p, true), g = [r, a], b = i != null, y = o != null;
  b && g.push(i), y && g.push(o);
  let v = [{ type: "int32", data: [m.padInfo.top, m.padInfo.left] }, { type: "int32", data: [m.strideHeight, m.strideWidth] }, { type: "int32", data: [m.dilationHeight, m.dilationWidth] }, { type: "int32", data: [m.inHeight, m.inWidth] }], x;
  return m.batchSize === 1 && m.inHeight === m.outHeight && m.inWidth === m.outWidth && m.strideHeight === 1 && m.strideWidth === 1 && m.filterHeight === m.filterWidth && m.inChannels === m.outChannels && m.dilationHeight === 1 && m.dilationWidth === 1 && m.filterHeight === 3 && m.inChannels % 4 === 0 ? x = new M2(m, b, d, y) : (x = new L2(m, b, d, y), v.push({ type: "int32", data: [m.filterHeight] }, { type: "int32", data: [m.filterWidth] }, { type: "int32", data: [m.outChannels / m.inChannels] })), d === "leakyrelu" && (v.push({ type: "float32", data: [h] }), x.uniforms += " alpha : f32,"), n.runWebGPUProgram(x, g, "float32", v);
}
var vae = { kernelName: oa, backendName: "webgpu", kernelFunc: yae };
var xae = class {
  constructor(e, t) {
    this.variableNames = ["A", "indices"], this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = t, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = `gathernd_${e}`, this.sliceDim = e, this.uniforms = `sliceDim : i32, strides : ${Ut(e)},`;
  }
  getUserCode() {
    let e;
    return this.sliceDim > 1 ? e = "uniforms.strides[j]" : e = "uniforms.strides", `
        ${Ue()}
        if (index < uniforms.size) {
          let coords = getCoordsFromIndex(index);
          var flattenIndex = 0;
          for (var j = 0; j < uniforms.sliceDim; j = j + 1) {
            let indexTemp = i32(round(getIndices(coords[0], j)));
            let strideNum = ${e};
            flattenIndex = flattenIndex + indexTemp * strideNum;
          }

          setOutputAtIndex(index, getA(flattenIndex, coords[1]));
        }
      }
      `;
  }
};
function wae(e) {
  let { inputs: t, backend: n } = e, { params: s, indices: r } = t, a = r.shape, i = a[a.length - 1], o = w.sizeFromShape(s.shape), [u, l, c, p] = C.prepareAndValidate(s, r), d = We({ inputs: { x: r }, backend: n, attrs: { shape: [l, i] } }), h = We({ inputs: { x: s }, backend: n, attrs: { shape: [w.sizeFromShape(s.shape) / c, c] } });
  if (n.shouldExecuteOnCPU([s, r]) || s.dtype === "string") {
    let y = n.readSync(r.dataId), v = n.bufferSync(s), x = cse(y, v, s.dtype, l, i, c, p, s.shape, o);
    return n.makeTensorInfo(u, s.dtype, x.values);
  }
  let f = new xae(i, [l, c]), m = [{ type: "int32", data: [i] }, { type: "int32", data: p }], g = n.runWebGPUProgram(f, [h, d], h.dtype, m), b = We({ inputs: { x: g }, backend: n, attrs: { shape: u } });
  return n.disposeData(d.dataId), n.disposeData(h.dataId), n.disposeData(g.dataId), b;
}
var kae = { kernelName: So, backendName: "webgpu", kernelFunc: wae };
var Sae = class {
  constructor(e, t) {
    this.variableNames = ["A", "indices"], this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e.slice(), this.aShape = e, this.outputShape = t, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = "gather";
  }
  getUserCode() {
    let e = Iae(this.aShape);
    return `
      ${Ue()}
        if (index < uniforms.size) {
          let resRC = getCoordsFromIndex(index);
          let indexZ = i32(getIndices(resRC.x, resRC.z));
          let inBounds = select(0.0, 1.0, indexZ >= 0 && indexZ < uniforms.aShape[2]);
          setOutputAtIndex(index, inBounds * getA(${e}));
        }
      }
    `;
  }
};
function Iae(e) {
  let t = ["resRC.x", "resRC.y", "resRC.z", "resRC.w"], n = [];
  for (let s = 0; s < e.length; s++)
    s === 2 ? n.push("indexZ") : n.push(`${t[s]}`);
  return n.join();
}
function W2(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r, indices: a } = t, { axis: i, batchDims: o } = s, u = w.parseAxisParam(i, r.shape)[0], l = C.segment_util.collectGatherOpShapeInfo(r, a, u, o), c = w.sizeFromShape(a.shape), p = [], d = We({ inputs: { x: r }, backend: n, attrs: { shape: [l.batchSize, l.outerSize, l.dimSize, l.sliceSize] } }), h = We({ inputs: { x: a }, backend: n, attrs: { shape: [l.batchSize, c / l.batchSize] } });
  p.push(d), p.push(h);
  let f = [l.batchSize, l.outerSize, c / l.batchSize, l.sliceSize];
  if (n.shouldExecuteOnCPU([r, a])) {
    let v = n.tensorMap.get(h.dataId).values, x = Ae(h.shape, h.dtype, v), I = n.tensorMap.get(d.dataId).values, $ = Ae(d.shape, d.dtype, I), R = dse($, x, f);
    return p.forEach((E) => n.disposeData(E.dataId)), n.makeTensorInfo(l.outputShape, R.dtype, R.values);
  }
  let m = new Sae(d.shape, f), g = n.runWebGPUProgram(m, [d, h], d.dtype);
  p.push(g);
  let b = We({ inputs: { x: g }, backend: n, attrs: { shape: l.outputShape } });
  return p.forEach((y) => n.disposeData(y.dataId)), b;
}
var Cae = { kernelName: ko, backendName: "webgpu", kernelFunc: W2 };
var Nae = mn({ opSnippet: 5, cpuKernelImpl: hse, dtype: "bool" });
var Tae = { kernelName: Io, backendName: "webgpu", kernelFunc: Nae };
var $ae = mn({ opSnippet: 6, dtype: "bool", cpuKernelImpl: pse });
var _ae = { kernelName: Va, backendName: "webgpu", kernelFunc: $ae };
function Aae(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { alpha: a } = s, i = [{ type: "float32", data: [a] }], o = new rc(r.shape, 14);
  return o.uniforms = "alpha : f32,", n.runWebGPUProgram(o, [r], "float32", i);
}
var Eae = { kernelName: Ua, backendName: "webgpu", kernelFunc: Aae };
var Rae = mn({ opSnippet: 7, dtype: "bool", cpuKernelImpl: mse });
var Dae = { kernelName: Co, backendName: "webgpu", kernelFunc: Rae };
var Fae = mn({ opSnippet: 8, dtype: "bool", cpuKernelImpl: fse });
var Oae = { kernelName: No, backendName: "webgpu", kernelFunc: Fae };
var Pae = Kt({ opType: 9, cpuKernelImpl: gse });
var zae = { kernelName: Ga, backendName: "webgpu", kernelFunc: Pae };
var Mae = mn({ opSnippet: 9, dtype: "bool" });
var Lae = { kernelName: To, backendName: "webgpu", kernelFunc: Mae };
var Bae = Kt({ opType: 10 });
var Vae = { kernelName: Sl, backendName: "webgpu", kernelFunc: Bae };
function U2(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { reductionIndices: a, keepDims: i } = s;
  return ic(r, a, i, "max", n);
}
var Wae = { kernelName: Ha, backendName: "webgpu", kernelFunc: U2 };
var Uae = mn({ opSnippet: 15, cpuKernelImpl: yse });
var Gae = { kernelName: qa, backendName: "webgpu", kernelFunc: Uae };
function Hae(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { filterSize: a, strides: i, pad: o, dimRoundingMode: u } = s, l = 1, c = C.computePool2DInfo(r.shape, a, i, l, o, u), p, d = [];
  if (c.filterHeight === 1 && c.filterWidth === 1) {
    if (w.arraysEqual(c.inShape, c.outShape))
      return Wn({ inputs: { x: r }, backend: n });
    p = new D2(c), d.push({ type: "int32", data: [c.strideHeight, c.strideWidth] });
  } else
    p = new R2(c, "max"), d.push({ type: "int32", data: [c.strideHeight, c.strideWidth] }, { type: "int32", data: [c.padInfo.top, c.padInfo.left] }, { type: "int32", data: [c.dilationHeight, c.dilationWidth] }, { type: "int32", data: [c.inHeight, c.inWidth] }, { type: "int32", data: [c.effectiveFilterHeight, c.effectiveFilterWidth] });
  return n.runWebGPUProgram(p, [r], r.dtype, d);
}
var qae = { kernelName: ja, backendName: "webgpu", kernelFunc: Hae };
function jae(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { keepDims: a, axis: i } = s;
  return ic(r, i, a, "mean", n);
}
var Kae = { kernelName: Ka, backendName: "webgpu", kernelFunc: jae };
function Xae(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  return ic(r, a, i, "min", n);
}
var Yae = { kernelName: Xa, backendName: "webgpu", kernelFunc: Xae };
var Qae = mn({ opSnippet: 16, cpuKernelImpl: vse });
var Zae = { kernelName: Ya, backendName: "webgpu", kernelFunc: Qae };
var Jae = class {
  constructor(e, t, n) {
    this.uniforms = "", this.variableNames = ["x"], this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = t.map((s, r) => s[0] + e[r] + s[1]), this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.xShape = e, t.map((s, r) => {
      this.uniforms += ` pad${r} : vec2<i32>,`;
    }), this.offset = n === "reflect" ? 0 : 1, this.shaderKey = `mirrorPad_${n}`;
  }
  getUserCode() {
    let e = this.xShape.length, t = this.xShape.map((u, l) => `uniforms.pad${l}[0]`).join(","), n = this.xShape.map((u, l) => `uniforms.pad${l}[0] + uniforms.xShape${e > 1 ? `[${l}]` : ""}`).join(","), s = e === 1 ? "start" : "start[i]", r = e === 1 ? "end" : "end[i]", a = e === 1 ? "outC" : "outC[i]", i = Ut(e), o = e > 1 ? ["coords[0]", "coords[1]", "coords[2]", "coords[3]"].slice(0, e) : "coords";
    return `
      ${Ue()}
        if (index < uniforms.size) {
          let start = ${i}(${t});
          let end = ${i}(${n});
          var outC = getCoordsFromIndex(index);
          for (var i = 0; i < ${e}; i = i + 1) {
            if (${a} < ${s}) {
              ${a} = ${s} * 2 - ${a} - ${this.offset};
            } else if(${a} >= ${r}) {
              ${a} = (${r} - 1) * 2 - ${a} + ${this.offset};
            }
          }
          let coords = outC - start;
          setOutputAtIndex(index, getX(${o}));
        }
      }
    `;
  }
};
var eie = { kernelName: Qa, backendName: "webgpu", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { x: s } = e, { paddings: r, mode: a } = t, i = n, o = r.map((c) => ({ type: "int32", data: [c[0], c[1]] })), u = new Jae(s.shape, r, a);
  return i.runWebGPUProgram(u, [s], s.dtype, o);
} };
function tie(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (n.shouldExecuteOnCPU([s])) {
    let a = n.tensorMap.get(s.dataId), [i, o] = wse(a.values, s.shape, s.dtype);
    return n.makeTensorInfo(o, s.dtype, i);
  }
  let r = new rc(s.shape, 11);
  return n.runWebGPUProgram(r, [s], s.dtype);
}
var nie = { kernelName: $o, backendName: "webgpu", kernelFunc: tie };
function sie(e) {
  console.warn("tf.nonMaxSuppression() in webgpu locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");
  let { inputs: t, backend: n, attrs: s } = e, { boxes: r, scores: a } = t, { maxOutputSize: i, iouThreshold: o, scoreThreshold: u } = s, l = n.readSync(r.dataId), c = n.readSync(a.dataId), { selectedIndices: p } = ws.nonMaxSuppressionV3Impl(l, c, i, o, u);
  return n.makeTensorInfo([p.length], "int32", new Int32Array(p));
}
var rie = { kernelName: Ao, backendName: "webgpu", kernelFunc: sie };
function aie(e) {
  console.warn("tf.nonMaxSuppression() in webgpu locks the UI thread. Call tf.nonMaxSuppressionAsync() instead");
  let { inputs: t, backend: n, attrs: s } = e, { boxes: r, scores: a } = t, { maxOutputSize: i, iouThreshold: o, scoreThreshold: u, softNmsSigma: l } = s, c = n.readSync(r.dataId), p = n.readSync(a.dataId), d = i, h = o, f = u, m = l, { selectedIndices: g, selectedScores: b } = ws.nonMaxSuppressionV5Impl(c, p, d, h, f, m);
  return [n.makeTensorInfo([g.length], "int32", new Int32Array(g)), n.makeTensorInfo([b.length], "float32", new Float32Array(b))];
}
var iie = { kernelName: Eo, backendName: "webgpu", kernelFunc: aie };
function qd(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (s.dtype === "complex64") {
    let r = ac({ inputs: { input: s }, backend: n }), a = qd({ inputs: { x: r }, backend: n }), i = ah({ inputs: { input: s }, backend: n }), o = qd({ inputs: { x: i }, backend: n }), u = hu({ inputs: { real: a, imag: o }, backend: n });
    return n.disposeData(r.dataId), n.disposeData(a.dataId), n.disposeData(i.dataId), n.disposeData(o.dataId), u;
  } else
    return mu({ attrs: { shape: s.shape, dtype: s.dtype, value: s.dtype === "string" ? "" : 0 }, backend: n });
}
var oie = { kernelName: Xo, backendName: "webgpu", kernelFunc: qd };
function G2(e) {
  let { inputs: t, backend: n } = e, { x: s } = t;
  if (s.dtype === "string")
    throw new Error("onesLike is not supported under string dtype");
  if (s.dtype === "complex64") {
    let r = ac({ inputs: { input: s }, backend: n }), a = G2({ inputs: { x: r }, backend: n }), i = ah({ inputs: { input: s }, backend: n }), o = qd({ inputs: { x: i }, backend: n }), u = hu({ inputs: { real: a, imag: o }, backend: n });
    return n.disposeData(r.dataId), n.disposeData(a.dataId), n.disposeData(i.dataId), n.disposeData(o.dataId), u;
  } else
    return mu({ attrs: { shape: s.shape, dtype: s.dtype, value: 1 }, backend: n });
}
var uie = { kernelName: Ro, backendName: "webgpu", kernelFunc: G2 };
function lie(e) {
  let { inputs: t, backend: n, attrs: s } = e, { axis: r } = s;
  if (t.length === 1)
    return sg({ inputs: { input: t[0] }, backend: n, attrs: { dim: r } });
  let a = t[0].shape, i = t[0].dtype;
  t.forEach((c) => {
    w.assertShapesMatch(a, c.shape, "All tensors passed to stack must have matching shapes"), w.assert(i === c.dtype, () => "All tensors passed to stack must have matching dtypes");
  });
  let o = [], u = t.map((c) => {
    let p = sg({ inputs: { input: c }, backend: n, attrs: { dim: r } });
    return o.push(p), p;
  }), l = O2({ inputs: u, backend: n, attrs: { axis: r } });
  return o.forEach((c) => n.disposeData(c.dataId)), l;
}
var cie = { kernelName: Fo, backendName: "webgpu", kernelFunc: lie };
var die = class {
  constructor(e, t) {
    this.variableNames = ["x"], this.uniforms = "constantValue : f32,", this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = t.map((n, s) => n[0] + e[s] + n[1]), this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), t.map((n, s) => {
      this.uniforms += ` pad${s} : vec2<i32>,`;
    }), this.xShape = e, this.shaderKey = "pad";
  }
  getUserCode() {
    let e = this.xShape.length, t = Ut(e), n = this.xShape.map((c, p) => `uniforms.pad${p}[0]`).join(","), s = this.xShape.map((c, p) => `uniforms.pad${p}[0] + uniforms.xShape${e > 1 ? `[${p}]` : ""}`).join(","), r = e > 1 ? `${t}(${n})` : `${n}`, a = e > 1 ? `${t}(${s})` : `${s}`, i = e > 1 ? "any(outC < start)" : "outC < start", o = e > 1 ? "any(outC >= end)" : "outC >= end", u = e > 1 ? ["coords[0]", "coords[1]", "coords[2]", "coords[3]"].slice(0, e) : "coords";
    return `
      ${Ue()}
        if (index < uniforms.size) {
          let start = ${r};
          let end = ${a};
          let outC = getCoordsFromIndex(index);

          if (${i} || ${o}) {
            setOutputAtIndex(index, uniforms.constantValue);
          } else {
            let coords = outC - start;
            setOutputAtIndex(index, getX(${u}));
          }
        }
      }
    `;
  }
};
var H2 = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { paddings: a, constantValue: i } = s;
  if (a.every((l) => w.arraysEqual(l, [0, 0])))
    return Wn({ inputs: { x: r }, backend: n });
  if (w.sizeFromShape(r.shape) === 0) {
    let l = a.map((c, p) => c[0] + r.shape[p] + c[1]);
    return mu({ backend: n, attrs: { shape: l, value: i, dtype: r.dtype } });
  }
  let o = [{ type: "float32", data: [i] }];
  a.map((l) => o.push({ type: "int32", data: [l[0], l[1]] }));
  let u = new die(r.shape, a);
  return n.runWebGPUProgram(u, [r], r.dtype, o);
};
var pie = { kernelName: Ja, backendName: "webgpu", kernelFunc: H2 };
var hie = mn({ opSnippet: 13 });
var fie = { kernelName: ei, backendName: "webgpu", kernelFunc: hie };
function mie(e) {
  let { inputs: t, backend: n } = e, { x: s, alpha: r } = t, a = new A2(14, s.shape, r.shape);
  return n.runWebGPUProgram(a, [s, r], "float32");
}
var gie = { kernelName: ti, backendName: "webgpu", kernelFunc: mie };
function bie(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, keepDims: i } = s;
  return ic(r, a, i, "prod", n);
}
var yie = { kernelName: ni, backendName: "webgpu", kernelFunc: bie };
var vie = (e) => {
  let { backend: t, attrs: n } = e, { start: s, stop: r, step: a, dtype: i } = n, o = Ise(s, r, a, i);
  return t.makeTensorInfo([o.length], i, o);
};
var xie = { kernelName: Nl, backendName: "webgpu", kernelFunc: vie };
var q2 = mn({ opSnippet: 3 });
var wie = { kernelName: Oa, backendName: "webgpu", kernelFunc: q2 };
var kie = Kt({ opType: 12 });
var Sie = { kernelName: si, backendName: "webgpu", kernelFunc: kie };
var Iie = Kt({ opType: 13 });
var Cie = { kernelName: ai, backendName: "webgpu", kernelFunc: Iie };
var Nie = class {
  constructor(e, t, n) {
    this.variableNames = ["x"], this.uniforms = "adjustHeightWidth : vec2<f32>, halfPixelCenters : f32,", this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = [e[0], t, n, e[3]], this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = "resizeBilinear";
  }
  getUserCode() {
    return `
      ${Ue()}
        if (index < uniforms.size) {
        let coords = getCoordsFromIndex(index);
          let b = coords[0];
          let d = coords[3];
          let rc = coords.yz;

          let effectiveInSize = vec2<f32>(
            f32(uniforms.xShape.y) - uniforms.adjustHeightWidth[0],
            f32(uniforms.xShape.z) - uniforms.adjustHeightWidth[1]);

          let effectiveOutSize = vec2<f32>(
            f32(uniforms.outShape.y) - uniforms.adjustHeightWidth[0],
            f32(uniforms.outShape.z) - uniforms.adjustHeightWidth[1]);

          let effectiveInputOverOutputRatioRC =
              effectiveInSize / effectiveOutSize;

          // Fractional source index
          let sourceFracIndexRC =
            (vec2<f32>(rc) + vec2<f32>(uniforms.halfPixelCenters)) *
            effectiveInputOverOutputRatioRC - vec2<f32>(uniforms.halfPixelCenters);

          // Compute the four integer indices.
          let sourceFloorRC = vec2<i32>(sourceFracIndexRC);
          let sourceCeilRC = vec2<i32>(
            min(vec2<f32>(uniforms.xShape.yz) - vec2<f32>(1.0), ceil(sourceFracIndexRC)));

          let topLeft = getX(b, sourceFloorRC.x, sourceFloorRC.y, d);
          let bottomLeft = getX(b, sourceCeilRC.x, sourceFloorRC.y, d);
          let topRight = getX(b, sourceFloorRC.x, sourceCeilRC.y, d);
          let bottomRight = getX(b, sourceCeilRC.x, sourceCeilRC.y, d);

          let fracRC = sourceFracIndexRC - vec2<f32>(sourceFloorRC);

          let top = topLeft + (topRight - topLeft) * fracRC.y;
          let bottom = bottomLeft + (bottomRight - bottomLeft) * fracRC.y;
          let newValue = top + (bottom - top) * fracRC.x;

          setOutputAtIndex(index, newValue);
        }
      }
    `;
  }
};
function Tie(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r } = t, { alignCorners: a, size: i, halfPixelCenters: o } = s, [u, l] = i, c = a && u > 1 ? 1 : 0, p = a && l > 1 ? 1 : 0, h = [{ type: "float32", data: [c, p] }, { type: "float32", data: [o ? 0.5 : 0] }], f = new Nie(r.shape, u, l);
  return n.runWebGPUProgram(f, [r], "float32", h);
}
var $ie = { kernelName: ri, backendName: "webgpu", kernelFunc: Tie };
var _ie = class {
  constructor(e, t, n, s) {
    this.variableNames = ["x"], this.uniforms = "adjustHeightWidth : vec2<f32>, roundBase : f32,", this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = [e[0], t, n, e[3]], this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.halfPixelCenters = s, this.shaderKey = `resizeNearest_${s}`;
  }
  getUserCode() {
    let e;
    return this.halfPixelCenters ? e = "max((vec2<f32>(rc) + vec2<f32>(0.5)) * effectiveInputOverOutputRatioRC, vec2<f32>(0.0))" : e = "vec2<f32>(rc) * effectiveInputOverOutputRatioRC", `
      ${Ue()}
        if (index < uniforms.size) {
          let coords = getCoordsFromIndex(index);
          let b = coords[0];
          let d = coords[3];
          let rc = coords.yz;

          let effectiveInSize = vec2<f32>(
            f32(uniforms.xShape.y) - uniforms.adjustHeightWidth[0],
            f32(uniforms.xShape.z) - uniforms.adjustHeightWidth[1]);

          let effectiveOutSize = vec2<f32>(
            f32(uniforms.outShape.y) - uniforms.adjustHeightWidth[0],
            f32(uniforms.outShape.z) - uniforms.adjustHeightWidth[1]);

          let effectiveInputOverOutputRatioRC =
              effectiveInSize / effectiveOutSize;

          // Fractional source index
          let sourceFracIndexRC = ${e};

          // Compute the coordinators of nearest neighbor point.
          let inputShapeRC = vec2<f32>(f32(uniforms.xShape.y), f32(uniforms.xShape.z));
          let sourceNearestRC = vec2<i32>(
            min(inputShapeRC - 1.0, floor(sourceFracIndexRC + uniforms.roundBase)));
          let newValue = getX(b, sourceNearestRC.x, sourceNearestRC.y, d);

          setOutputAtIndex(index, newValue);
        }
      }
    `;
  }
};
function Aie(e) {
  let { inputs: t, backend: n, attrs: s } = e, { images: r } = t, { alignCorners: a, halfPixelCenters: i, size: o } = s, [u, l] = o, c = a && u > 1 ? 1 : 0, p = a && l > 1 ? 1 : 0, h = [{ type: "float32", data: [c, p] }, { type: "float32", data: [a ? 0.5 : 0] }], f = new _ie(r.shape, u, l, i);
  return n.runWebGPUProgram(f, [r], r.dtype, h);
}
var Eie = { kernelName: $l, backendName: "webgpu", kernelFunc: Aie };
var Rie = class {
  constructor(e, t) {
    this.outputShape = [], this.variableNames = ["x"], this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.uniforms = `centerX : f32, centerY : f32, sinRadians : f32,
          cosRadians : f32,`, this.shaderKey = "rotate", this.outputShape = e, typeof t == "number" ? (this.uniforms += " fillValue : f32,", this.fillSnippet = "var outputValue = uniforms.fillValue;", this.shaderKey += "_float") : (this.uniforms += " fillValue : vec3<f32>,", this.fillSnippet = "var outputValue = uniforms.fillValue[coords[3]];", this.shaderKey += "_vec3");
  }
  getUserCode() {
    return `
        ${Ue()}

          if (index < uniforms.size) {
            let coords = getCoordsFromIndex(index);
            let coordXFloat = (f32(coords[2]) - uniforms.centerX) *
                uniforms.cosRadians - (f32(coords[1]) - uniforms.centerY) *
                uniforms.sinRadians;
            let coordYFloat = (f32(coords[2]) - uniforms.centerX) *
                uniforms.sinRadians + (f32(coords[1]) - uniforms.centerY) *
                uniforms.cosRadians;
            let coordX = i32(round(coordXFloat + uniforms.centerX));
            let coordY = i32(round(coordYFloat + uniforms.centerY));
            ${this.fillSnippet}
            if(coordX >= 0 && coordX < uniforms.xShape[2] && coordY >= 0 &&
                coordY < uniforms.xShape[1]) {
              outputValue = getX(coords[0], coordY, coordX, coords[3]);
            }
            setOutputAtIndex(index, outputValue);
          }
        }
      `;
  }
};
var Die = { kernelName: Yo, backendName: "webgpu", kernelFunc: ({ inputs: e, attrs: t, backend: n }) => {
  let { image: s } = e, { radians: r, fillValue: a, center: i } = t, o = n, u = new Rie(s.shape, a), [l, c] = C.getImageCenter(i, s.shape[1], s.shape[2]), p = [{ type: "float32", data: [l] }, { type: "float32", data: [c] }, { type: "float32", data: [Math.sin(r)] }, { type: "float32", data: [Math.cos(r)] }];
  return typeof a == "number" ? p.push({ type: "float32", data: [Number.parseFloat(a.toFixed(2))] }) : p.push({ type: "float32", data: a }), o.runWebGPUProgram(u, [s], s.dtype, p);
} };
var Fie = Kt({ opType: 15, cpuKernelImpl: Cse });
var Oie = { kernelName: ii, backendName: "webgpu", kernelFunc: Fie };
var Pie = class {
  constructor(e, t, n, s, r, a, i) {
    this.variableNames = ["updates", "indices"], this.workGroupSize = [64, 1, 1], this.atomic = true, this.outputShape = a, this.type = i, this.dispatchLayout = Be(e), this.dispatch = _e(this.dispatchLayout, e, this.workGroupSize), this.sliceDimGreaterThanOne = t > 1, this.shaderKey = `scatter_${n}_${s}_${this.sliceDimGreaterThanOne}_${i}`;
    let o = Ut(r.length);
    this.uniforms = `sliceDim : i32, strides: ${o}, size: i32,`, this.updatesRank = s, this.indicesRank = n;
  }
  getUserCode() {
    let e = "";
    this.indicesRank === 1 ? e = "coords[0]" : this.indicesRank === 2 && (e = "coords[0], j");
    let t = `getIndices(${e})`, n = this.sliceDimGreaterThanOne ? "uniforms.strides[j]" : "uniforms.strides", s = "", r = "", a = "";
    this.updatesRank === 1 ? (s = "coords[0]", r = "flattenedIndex", a = `
      fn getUpdatesCoordsFromFlatIndex(index : i32) -> i32 {
        return index;
      }
      `) : this.updatesRank === 2 && (s = "coords[0], coords[1]", r = "vec2<i32>(flattenedIndex, coords[1])", a = `
      fn getUpdatesCoordsFromFlatIndex(index : i32) -> vec2<i32> {
        let d0 = index / uniforms.updatesShape[1];
        let d1 = index - d0 * uniforms.updatesShape[1];
        return vec2<i32>(d0, d1);
      }
      `);
    let i = `getUpdates(${s})`, o = this.type === "int32" ? "atomicAdd(&(result[flatIndex]), i32(updateValue));" : `
     var assumed = atomicLoad(&(result[flatIndex]));
     var success = 0;
     for (; success == 0;) {
       let new = bitcast<f32>(assumed) + updateValue;
       let newI32 = bitcast<i32>(new);
       let resValue = atomicCompareExchangeWeak(&(result[flatIndex]), assumed, newI32);
       assumed = resValue[0];
       success = resValue[1];
     }
     `;
    return `
    ${a}

      ${Ue()}

        if (index < uniforms.size) {
          let coords = getUpdatesCoordsFromFlatIndex(index);
          var flattenedIndex = 0;
          for (var j = 0; j < uniforms.sliceDim; j = j + 1) {
            let indexInside = i32(round(${t}));
            flattenedIndex = flattenedIndex + indexInside * ${n};
          }
          let updateValue = ${i};
          let flatIndex = getOutputIndexFromCoords(${r});

         ${o}
        }
      }`;
  }
};
function zie(e) {
  let { inputs: t, backend: n, attrs: s } = e, { indices: r, updates: a } = t, { shape: i } = s, { sliceRank: o, numUpdates: u, sliceSize: l, strides: c, outputSize: p } = C.calculateShapes(a, r, i), d = [p / l, l];
  if (p === 0)
    return n.makeTensorInfo(i, r.dtype);
  let h = We({ inputs: { x: r }, backend: n, attrs: { shape: [u, o] } }), f = We({ inputs: { x: a }, backend: n, attrs: { shape: [u, l] } }), m = f.dtype, g = mu({ backend: n, attrs: { shape: d, value: 0, dtype: m } }), b = w.sizeFromShape(f.shape), y = [{ type: "int32", data: [o] }, { type: "int32", data: c }, { type: "int32", data: [b] }], v = new Pie(f.shape, o, h.shape.length, f.shape.length, c, d, m), x = n.runWebGPUProgram(v, [f, h], m, y, g), k = We({ inputs: { x }, backend: n, attrs: { shape: i } });
  return n.disposeData(h.dataId), n.disposeData(f.dataId), n.disposeData(x.dataId), k;
}
var Mie = { kernelName: Mo, backendName: "webgpu", kernelFunc: zie };
var Lie = class {
  constructor(e, t, n) {
    this.variableNames = ["c", "a", "b"], this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = t, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.cRank = e, this.rank = n, this.shaderKey = "select";
  }
  getUserCode() {
    let e, t;
    if (this.rank > 4)
      throw Error(`Where for rank ${this.rank} is not yet supported`);
    if (this.rank === 1)
      t = "resRC", e = "resRC";
    else {
      let s = ["resRC.x", "resRC.y", "resRC.z", "resRC.w"], r = [], a = [];
      for (let i = 0; i < this.outputShape.length; i++)
        a.push(`${s[i]}`), i < this.cRank && r.push(`${s[i]}`);
      e = r.join(), t = a.join();
    }
    return `
      ${Ue()}
        if (index < uniforms.size) {
          let resRC = getCoordsFromIndex(index);
          let cVal = getC(${e});
          if (cVal >= 1.0) {
            setOutputAtIndex(index, getA(${t}));
          } else {
            setOutputAtIndex(index, getB(${t}));
          }
        }
      }
    `;
  }
};
function Bie(e) {
  let { inputs: t, backend: n } = e, { condition: s, t: r, e: a } = t, i = new Lie(s.shape.length, r.shape, r.shape.length);
  return n.runWebGPUProgram(i, [s, r, a], cn(r.dtype, a.dtype));
}
var Vie = { kernelName: Lo, backendName: "webgpu", kernelFunc: Bie };
var Wie = Kt({ opType: 18 });
var Uie = { kernelName: ui, backendName: "webgpu", kernelFunc: Wie };
var Gie = Kt({ opType: 16 });
var Hie = { kernelName: oi, backendName: "webgpu", kernelFunc: Gie };
var qie = Kt({ opType: 17 });
var jie = { kernelName: Vo, backendName: "webgpu", kernelFunc: qie };
var j2 = mn({ opSnippet: 2, cpuKernelImpl: Ese, supportsComplex: true });
var Kie = { kernelName: hi, backendName: "webgpu", kernelFunc: j2 };
function Xie(e) {
  let { inputs: t, backend: n, attrs: s } = e, { logits: r } = t, { dim: a } = s, i = w.parseAxisParam([a], r.shape), o = U2({ inputs: { x: r }, backend: n, attrs: { reductionIndices: i, keepDims: false } }), u = C.expandShapeToKeepDim(o.shape, i), l = We({ inputs: { x: o }, backend: n, attrs: { shape: u } }), c = j2({ inputs: { a: r, b: l }, backend: n }), p = V2({ inputs: { x: c }, backend: n }), d = zv({ inputs: { x: p }, backend: n, attrs: { axis: i, keepDims: false } }), h = We({ inputs: { x: d }, backend: n, attrs: { shape: u } }), f = q2({ inputs: { a: p, b: h }, backend: n });
  return n.disposeData(o.dataId), n.disposeData(l.dataId), n.disposeData(c.dataId), n.disposeData(p.dataId), n.disposeData(d.dataId), n.disposeData(h.dataId), f;
}
var Yie = { kernelName: di, backendName: "webgpu", kernelFunc: Xie };
var Qie = (e) => {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockShape: a, paddings: i } = s;
  w.assert(r.shape.length <= 4, () => "spaceToBatchND for rank > 4 with a WebGPU backend not implemented yet");
  let o = a.reduce((b, y) => b * y), u = [[0, 0]];
  u.push(...i);
  for (let b = 1 + a.length; b < r.shape.length; ++b)
    u.push([0, 0]);
  let l = [], c = H2({ inputs: { x: r }, backend: n, attrs: { paddings: u, constantValue: 0 } }), p = C.getReshaped(c.shape, a, o, false), d = C.getPermuted(p.length, a.length, false), h = C.getReshapedPermuted(c.shape, a, o, false), f = We({ inputs: { x: c }, backend: n, attrs: { shape: p } }), m = Ks({ inputs: { x: f }, backend: n, attrs: { perm: d } }), g = We({ inputs: { x: m }, backend: n, attrs: { shape: h } });
  return l.push(c), l.push(f), l.push(m), l.forEach((b) => n.disposeData(b.dataId)), g;
};
var Zie = { kernelName: Wo, backendName: "webgpu", kernelFunc: Qie };
var Jie = class {
  constructor(e, t, n, s, r, a, i = true) {
    this.variableNames = ["updates", "indices", "defaultValue"], this.workGroupSize = [64, 1, 1], this.workPerThread = 4, this.size = true, this.outputShape = a, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]);
    let o = t > 1;
    this.shaderKey = `scatter_${n}_${s}_${o}`;
    let u = Ut(r.length);
    this.uniforms = `updateSize : i32, sliceDim : i32, strides: ${u},`;
    let l = "";
    n === 1 ? l = "i" : n === 2 && (l = "i, j"), this.indicesSnippet = `getIndices(${l})`;
    let c = "";
    s === 1 ? c = "i" : s === 2 && (c = "i, coords[1]"), this.updatesSnippet = `getUpdates(${c})`, this.strideString = o ? "uniforms.strides[j]" : "uniforms.strides";
  }
  getUserCode() {
    return `
      ${Ue()}

        let globalIndex = index * ${this.workPerThread};
        if (globalIndex < uniforms.size) {
          var sum = vec4<f32>(0.0);
          var found = vec4<bool>(false);
          for (var i = 0; i < uniforms.updateSize; i = i + 1) {
            var flattenedIndex = 0;
            for (var j = 0; j < uniforms.sliceDim; j = j + 1) {
              let indexInside = i32(round(${this.indicesSnippet}));
              flattenedIndex = flattenedIndex + indexInside * ${this.strideString};
            }
            for (var innerIndex = 0; innerIndex < ${this.workPerThread}; innerIndex = innerIndex + 1) {
              let curIndex = globalIndex + innerIndex;
              let coords = getCoordsFromIndex(curIndex);
              if (flattenedIndex == coords[0]) {
                sum[innerIndex] = sum[innerIndex] + ${this.updatesSnippet};
                found[innerIndex] = true;
              }
            }
          }
          for (var innerIndex = 0; innerIndex < ${this.workPerThread}; innerIndex = innerIndex + 1) {
            let curIndex = globalIndex + innerIndex;
            if (curIndex < uniforms.size)
            {
              setOutputAtIndex(curIndex, mix(getDefaultValue(), sum[innerIndex], f32(found[innerIndex])));
            }
          }
        }
      }`;
  }
};
function eoe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { sparseIndices: r, sparseValues: a, defaultValue: i } = t, { outputShape: o } = s, { sliceRank: u, numUpdates: l, sliceSize: c, strides: p, outputSize: d } = C.calculateShapes(a, r, o), h = false;
  if (a.dtype === "string") {
    let y = n.bufferSync(r), v = n.bufferSync(a), x = w.decodeString(n.readSync(i.dataId)[0]), k = Nse(y, v, o, d, c, l, u, p, x, h);
    return n.makeTensorInfo(o, k.dtype, k.values);
  }
  let f = [{ type: "int32", data: [l] }, { type: "int32", data: [u] }, { type: "int32", data: p }], m = new Jie(l, u, r.shape.length, a.shape.length, p, [d, 1], h), g = n.runWebGPUProgram(m, [a, r, i], a.dtype, f), b = We({ inputs: { x: g }, backend: n, attrs: { shape: o } });
  return n.disposeData(g.dataId), b;
}
var toe = { kernelName: hp, backendName: "webgpu", kernelFunc: eoe };
function noe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { numOrSizeSplits: a, axis: i } = s, o = w.parseAxisParam(i, r.shape)[0], u = C.prepareSplitSize(r, a, o), l = r.shape.length, c = new Array(l).fill(0), p = r.shape.slice();
  return u.map((d) => {
    let h = [...p];
    h[o] = d;
    let f = fu({ inputs: { x: r }, backend: n, attrs: { begin: c, size: h } });
    return c[o] += d, f;
  });
}
var soe = { kernelName: Uo, backendName: "webgpu", kernelFunc: noe };
var roe = Kt({ opType: 19 });
var aoe = { kernelName: li, backendName: "webgpu", kernelFunc: roe };
var ioe = { kernelName: Dl, backendName: "webgpu", kernelFunc: ({ inputs: e, backend: t }) => {
  let { x: n } = e, s = t, r = new rc(n.shape, 20);
  return s.runWebGPUProgram(r, [n], n.dtype);
} };
var ooe = mn({ opSnippet: 11 });
var uoe = { kernelName: pi, backendName: "webgpu", kernelFunc: ooe };
var loe = class {
  constructor(e) {
    this.variableNames = ["x"], this.workPerThread = 1, this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize, [this.workPerThread, 1, 1]);
    let t = Ut(this.outputShape.length);
    this.uniforms = `begin : ${t},  strides : ${t}, `, this.shaderKey = "stridedSlice";
  }
  getUserCode() {
    let e = this.outputShape.length, t = "";
    if (e === 1)
      t = "coords * uniforms.strides + uniforms.begin";
    else {
      let s = 0;
      t = this.outputShape.map((r, a) => (s++, this.outputShape.length === 1 ? `coords * uniforms.strides[${a}] + uniforms.begin[${a}]` : `coords[${s - 1}] * uniforms.strides[${a}] + uniforms.begin[${a}]`)).join(",");
    }
    return `
       ${Ue()}
         if (index < uniforms.size) {
           let coords = getCoordsFromIndex(index);
           setOutputAtIndex(index, getX(${t}));
         }
       }
     `;
  }
};
function coe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { begin: a, end: i, strides: o, beginMask: u, endMask: l, ellipsisMask: c, newAxisMask: p, shrinkAxisMask: d } = s, { finalShapeSparse: h, finalShape: f, isIdentity: m, sliceDim0: g, isSimpleSlice: b, begin: y, end: v, strides: x } = wt.sliceInfo(r.shape, a, i, o, u, l, c, p, d), k;
  if (m)
    k = We({ inputs: { x: r }, backend: n, attrs: { shape: f } });
  else if (g || b) {
    w.assert(r.shape.length >= 1, () => `Input must have rank at least 1, got: ${r.shape.length}`);
    let I = wt.computeOutShape(y, v, x), $ = fu({ inputs: { x: r }, backend: n, attrs: { begin: y, size: I } });
    k = We({ inputs: { x: $ }, backend: n, attrs: { shape: f } }), n.disposeData($.dataId);
  } else if (n.shouldExecuteOnCPU([r])) {
    let $ = n.readSync(r.dataId), R = Ae(r.shape, r.dtype, $), E = _se(h, R, x, y);
    k = n.makeTensorInfo(f, r.dtype, E.values);
  } else {
    let $ = new loe(h), R = [{ type: "int32", data: y }, { type: "int32", data: x }], E = n.runWebGPUProgram($, [r], r.dtype, R);
    k = We({ inputs: { x: E }, backend: n, attrs: { shape: f } }), n.disposeData(E.dataId);
  }
  return k;
}
var doe = { kernelName: Go, backendName: "webgpu", kernelFunc: coe };
function poe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { separator: r, nGramWidths: a, leftPad: i, rightPad: o, padWidth: u, preserveShortSequences: l } = s, { data: c, dataSplits: p } = t, d = n.readSync(c.dataId), h = n.readSync(p.dataId), [f, m] = Ase(d, h, r, a, i, o, u, l);
  return [n.makeTensorInfo([f.length], "string", f), n.makeTensorInfo(p.shape, "int32", m)];
}
var hoe = { kernelName: fp, backendName: "webgpu", kernelFunc: poe };
var foe = Kt({ opType: 21 });
var moe = { kernelName: fi, backendName: "webgpu", kernelFunc: foe };
var goe = class {
  constructor(e, t) {
    this.variableNames = ["A"], this.workGroupSize = [64, 1, 1], this.size = true;
    let n = new Array(e.length);
    for (let s = 0; s < n.length; s++)
      n[s] = e[s] * t[s];
    this.outputShape = n, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.rank = this.outputShape.length, this.shaderKey = "tile";
  }
  getUserCode() {
    let e = boe(this.rank, "uniforms.");
    return `
      ${Ue()}
        if (index < uniforms.size) {
          let resRC = getCoordsFromIndex(index);
          setOutputAtIndex(index, getA(${e}));
        }
      }
    `;
  }
};
function boe(e, t = "") {
  if (e >= 5)
    throw Error(`Tile for rank ${e} is not yet supported`);
  if (e === 1)
    return `(resRC % ${t}aShape)`;
  let n = ["resRC.x", "resRC.y", "resRC.z", "resRC.w"], s = [];
  for (let r = 0; r < e; r++)
    s.push(`(${n[r]} % ${t}aShape[${r}])`);
  return s.join();
}
function yoe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { reps: a } = s;
  if (n.shouldExecuteOnCPU([r]) || r.dtype === "string" || r.shape.length >= 5) {
    let u = n.readSync(r.dataId), l = r.dtype === "string" ? u.map((d) => w.decodeString(d)) : u, c = Ae(r.shape, r.dtype, l), p = Rse(c, a);
    return n.makeTensorInfo(p.shape, p.dtype, p.values);
  }
  let i = new goe(r.shape, a);
  return n.runWebGPUProgram(i, [r], r.dtype);
}
var voe = { kernelName: Cr, backendName: "webgpu", kernelFunc: yoe };
var xoe = class {
  constructor(e) {
    this.variableNames = ["x", "indices"], this.workGroupSize = [256, 1, 1], this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.uniforms = `inputSize : i32, firstPass : i32, negativeInf : f32,
        dir : i32, inc : i32,`, this.shaderKey = "swap";
  }
  getUserCode() {
    return `
        ${Ue()}
          if (index < uniforms.size) {
            let outC = getCoordsFromIndex(index);
            let batch = outC[0];
            let elemIdx = outC[1];
            // We compare elements pair-wise within a group of size 2 * inc.
            // The comparing rule for each group alternates between ascending
            // and descending. Within each group, we compare each pair at
            // positions i and i+inc. To decide whether an element at position i
            // is x0 or x1, we mod it by 2 * inc, if the result is smaller than
            // inc, it is in the first half of the group, we denote it as x0,
            // otherwise we denote it as x1.
            // For example, as shown in the Bitonic top K paper referenced
            // above, Figure5(a) shows that element[1] is in the second half of
            // the group when group size is 2, but it is in the first half of
            // the group when group size is 4.
            let isFirstInPair = elemIdx % (2 * uniforms.inc) < uniforms.inc;
            var i = 0;
            if (isFirstInPair) {
              i = elemIdx;
            } else {
              i = elemIdx - uniforms.inc;
            }

            var i0 = 0;
            if (uniforms.firstPass == 1) {
              i0 = i;
            } else {
              i0 = i32(getIndices(batch, i));
            }

            var i1 = 0;
            if (uniforms.firstPass == 1) {
              i1 = i + uniforms.inc;
            } else {
              i1 = i32(getIndices(batch, i + uniforms.inc));
            }

            var x0 = f32(0.0);
            var x1 = f32(0.0);
            if (i0 < uniforms.inputSize) {
              x0 = getX(batch, i0);
            } else {
              x0 = uniforms.negativeInf;
            }
            if (i1 < uniforms.inputSize) {
              x1 = getX(batch, i1);
            } else {
              x1 = uniforms.negativeInf;
            }

            let reverse = elemIdx % (2 * uniforms.dir) >= uniforms.dir;
            let isGreater = x0 > x1 || (x0 == x1 && i1 > i0);
            if (reverse == isGreater) {
              // Elements in opposite order of direction
              let iTemp = i0;
              i0 = i1;
              i1 = iTemp;
            }
            if (isFirstInPair) {
              setOutputAtIndex(index, f32(i0));
            } else {
              setOutputAtIndex(index, f32(i1));
            }
          }
        }
      `;
  }
};
var woe = class {
  constructor(e) {
    this.variableNames = ["x", "indices"], this.workGroupSize = [256, 1, 1], this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.uniforms = "inputSize : i32, firstPass : i32, k : i32,", this.shaderKey = "merge";
  }
  getUserCode() {
    return `
        ${Ue()}
          if (index < uniforms.size) {
            let outC = getCoordsFromIndex(index);
            let batch = outC[0];
            let elemIdx = outC[1];
            // The output size is half of the previous size.
            // If the previous sequence is | | | | _ _ _ _  | | | |  _ _ _ _
            // (k=4), we only need to output the indices at positions |, the
            // indices at positions _ can be thrown away, see Figure5(b) After
            // Phase 2 (Merge phase) in the Bitonic Top K paper referenced
            // above.
            // For example, the paper shows we only need to output the orange
            // bars. The output sequence should look like this | | | | | | | |.
            // Because the sequence is halved, to map the output index back to
            // the previous sequence to find the corresponding value, we need
            // to double the index. When we double the index, we basically
            // interpolate a position, so 2i looks like
            // | _ | _ | _ | _ | _ | _ | _. We move the | to the first k
            // position of each 2k positions by - elemIdx % k. E.g. for output
            // at index 4,5,6,7, we want to get the corresponding element at
            // original index 8,9,10,11, for output at index 8,9,10,11,
            // we want to get the corresponding element at original index
            // 16,17,18,19, so on and so forth.

            var i = 0;
            if (elemIdx < uniforms.k) {
              i = elemIdx;
            } else {
              i = elemIdx * 2 - elemIdx % uniforms.k;
            }
            var i0 = 0;
            if (uniforms.firstPass == 1) {
              i0 = i;
            } else {
              i0 = i32(getIndices(batch, i));
            }
            var i1 = 0;
            if (uniforms.firstPass == 1) {
              i1 = i + uniforms.k;
            } else {
              i1 = i32(getIndices(batch, i + uniforms.k));
            }

            let x0 = getX(batch, i0);
            var x1 = f32(0.0);
            if (i1 < uniforms.inputSize) {
              x1 = getX(batch, i1);
            } else {
              x1 = x0;
            }

            if (x0 >= x1) {
              setOutputAtIndex(index, f32(i0));
            } else {
              setOutputAtIndex(index, f32(i1));
            }
          }
        }
      `;
  }
};
function Gi(e, t) {
  t !== null && e.disposeData(t.dataId);
}
function Ww(e) {
  let t = 1;
  for (; t < e; )
    t *= 2;
  return t;
}
function koe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { k: a, sorted: i } = s, o = r.shape, u = o[o.length - 1];
  if (n.shouldExecuteOnCPU([r])) {
    let k = n.readSync(r.dataId), [I, $] = Dse(k, o, r.dtype, a, i);
    return [n.makeTensorInfo(I.shape, I.dtype, I.values), n.makeTensorInfo($.shape, $.dtype, $.values)];
  }
  if (a === 0)
    return o[o.length - 1] = 0, [n.makeTensorInfo(o, r.dtype, []), n.makeTensorInfo(o, "int32", [])];
  if (u === 1)
    return [r, mu({ attrs: { shape: o, dtype: "int32", value: 0 }, backend: n })];
  let c = w.sizeFromShape(o) / u, p = We({ inputs: { x: r }, attrs: { shape: [c, u] }, backend: n }), d = Ww(a), h = Ww(u), f = null, m = () => f === null ? [p, p] : [p, f], g = (k, I, $) => {
    let R = m(), E = new xoe($), A = [{ type: "int32", data: [u] }, { type: "int32", data: [f === null ? 1 : 0] }, { type: "float32", data: [Number.NEGATIVE_INFINITY] }, { type: "int32", data: [k] }, { type: "int32", data: [I] }], O = f;
    f = n.runWebGPUProgram(E, R, "int32", A), Gi(n, O);
  };
  for (let k = 1; k < d; k *= 2) {
    let I = k * 2;
    for (let $ = k; $ >= 1; $ /= 2)
      g(I, $, [c, h]);
  }
  for (let k = h; k > d; k /= 2) {
    let I = m(), $ = new woe([c, k / 2]), E = [{ type: "int32", data: [u] }, { type: "int32", data: [f === null ? 1 : 0] }, { type: "int32", data: [d] }], P = f;
    f = n.runWebGPUProgram($, I, "int32", E), Gi(n, P);
    let A = d / 2, O = A * 2;
    for (let T = A; T >= 1; T /= 2)
      g(O, T, f.shape);
  }
  let b = f;
  f = fu({ inputs: { x: f }, backend: n, attrs: { begin: 0, size: [c, a] } }), Gi(n, b);
  let y = W2({ inputs: { x: p, indices: f }, backend: n, attrs: { axis: 1, batchDims: 1 } });
  Gi(n, p);
  let v = o.slice(0, -1);
  v.push(a), b = f, f = We({ inputs: { x: f }, attrs: { shape: v }, backend: n }), Gi(n, b);
  let x = y;
  return y = We({ inputs: { x: y }, attrs: { shape: v }, backend: n }), Gi(n, x), [y, f];
}
var Soe = { kernelName: qo, backendName: "webgpu", kernelFunc: koe };
var Ioe = class {
  constructor(e) {
    this.variableNames = ["Image", "Transforms"], this.uniforms = "interpolationModeId : i32, fillModeId : i32, fillValue : f32,", this.workGroupSize = [64, 1, 1], this.size = true, this.outputShape = e, this.dispatchLayout = Be(this.outputShape), this.dispatch = _e(this.dispatchLayout, this.outputShape, this.workGroupSize), this.shaderKey = "transform";
  }
  getUserCode() {
    return `
          fn mapCoord(outCoord : f32, len : f32) -> f32{
            var inCoord = outCoord;
            if(uniforms.fillModeId == 2) {
              if (inCoord < 0.0) {
                if (len <= 1.0) {
                  inCoord = 0.0;
                } else {
                  let sz2 = 2.0 * len;
                  if (inCoord < sz2) {
                    inCoord = sz2 * f32(i32(f32(-inCoord / sz2))) +
                    inCoord;
                  }
                  if (inCoord < -len) {
                    inCoord = inCoord + sz2;
                  } else {
                    inCoord = -inCoord - 1.0;
                  }
                }
              } else if (inCoord > len - 1.0) {
                if (len <= 1.0) {
                  inCoord = 0.0;
                } else {
                  let sz2 = 2.0 * len;
                  inCoord = inCoord - sz2 * f32(i32(f32(inCoord / sz2)));
                  if (inCoord >= len) {
                    inCoord = sz2 - inCoord - 1.0;
                  }
                }
              }
              return clamp(inCoord, 0.0, len - 1.0);
            } else if (uniforms.fillModeId == 3) {
              if (inCoord < 0.0) {
                if (len <= 1.0) {
                  inCoord = 0.0;
                } else {
                  let sz = len - 1.0;
                  inCoord = inCoord + len * (f32(i32(f32(-inCoord / sz))) + 1.0);
                }
              } else if (inCoord > len - 1.0) {
                if (len <= 1.0) {
                  inCoord = 0.0;
                } else {
                  let sz = len - 1.0;
                  inCoord = inCoord - len * f32(i32(f32(inCoord / sz)));
                }
              }
              return clamp(inCoord, 0.0, len - 1.0);
            } else if (uniforms.fillModeId == 4) {
              return clamp(outCoord, 0.0, len - 1.0);
            }
            return outCoord;
          }
          fn readWithFillValue(batch : i32, coordY : i32, coordX : i32,
            channel : i32) -> f32 {
            var outputValue : f32;
            if (0 <= coordY && coordY < uniforms.imageShape[1] && 0 <= coordX && coordX < uniforms.imageShape[2]) {
                outputValue = getImage(batch, coordY, coordX, channel);
            } else {
              outputValue = uniforms.fillValue;
            }
            return outputValue;
          }

          ${Ue()}
            if (index < uniforms.size) {
              let coords = getCoordsFromIndex(index);
              var outputValue : f32;
              let batch = coords[0];
              let x = coords[2];
              let y = coords[1];
              let channel = coords[3];
              let xf = f32(x);
              let yf = f32(y);
              let a1 = getTransforms(batch, 0);
              let a2 = getTransforms(batch, 1);
              let a3 = getTransforms(batch, 2);
              let b1 = getTransforms(batch, 3);
              let b2 = getTransforms(batch, 4);
              let b3 = getTransforms(batch, 5);
              let c1 = getTransforms(batch, 6);
              let c2 = getTransforms(batch, 7);
              let projection = c1 * xf + c2 * yf + 1.0;
              if (projection == 0.0) {
                outputValue = uniforms.fillValue;
              } else {
                let inX = (a1 * xf + a2 * yf + a3) / projection;
                let inY = (b1 * xf + b2 * yf + b3) / projection;
                let mapX = mapCoord(inX, f32(uniforms.imageShape[2]));
                let mapY = mapCoord(inY, f32(uniforms.imageShape[1]));

                if (uniforms.interpolationModeId == 1) {
                  let coordY = i32(round(mapY));
                  let coordX = i32(round(mapX));
                  outputValue = readWithFillValue(batch, coordY, coordX,
                    channel);
                } else {
                  let yFloor = floor(mapY);
                  let xFloor = floor(mapX);
                  let yCeil = yFloor + 1.0;
                  let xCeil = xFloor + 1.0;
                  let valueYFloor = (xCeil - mapX) *
                  readWithFillValue(batch, i32(yFloor), i32(xFloor), channel) +
                  (mapX - xFloor) *
                  readWithFillValue(batch, i32(yFloor), i32(xCeil), channel);
                  let valueYCeil = (xCeil - mapX) *
                  readWithFillValue(batch, i32(yCeil), i32(xFloor), channel) +
                  (mapX - xFloor) *
                  readWithFillValue(batch, i32(yCeil), i32(xCeil), channel);
                  outputValue = (yCeil - mapY) * valueYFloor +
                  (mapY - yFloor) * valueYCeil;
                }
              }
              setOutputAtIndex(index, outputValue);
            }
          }
        `;
  }
};
function Coe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { image: r, transforms: a } = t, { interpolation: i, fillMode: o, fillValue: u, outputShape: l } = s, [c, p, d, h] = r.shape, [f, m] = l != null ? l : [p, d], g = [c, f, m, h], b = new Ioe(g), y = i === "nearest" ? 1 : 2, v;
  switch (o) {
    case "constant":
      v = 1;
      break;
    case "reflect":
      v = 2;
      break;
    case "wrap":
      v = 3;
      break;
    case "nearest":
      v = 4;
      break;
    default:
      v = 1;
      break;
  }
  let x = [{ type: "int32", data: [y] }, { type: "int32", data: [v] }, { type: "float32", data: [u] }];
  return n.runWebGPUProgram(b, [r, a], "float32", x);
}
var Noe = { kernelName: jo, backendName: "webgpu", kernelFunc: Coe };
function Toe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { value: r } = t, { axis: a } = s;
  a < 0 && (a += r.shape.length);
  let i = r, o = i.shape.length, u = r.shape[a], l = new Array(o - 1), c = 0;
  for (let m = 0; m < o; m++)
    m !== a && (l[c++] = i.shape[m]);
  let p = [], d = new Array(o).fill(0), h = i.shape.slice();
  h[a] = 1;
  let f = new Array(u);
  for (let m = 0; m < f.length; m++) {
    d[a] = m;
    let g = fu({ inputs: { x: i }, backend: n, attrs: { begin: d, size: h } }), b = We({ inputs: { x: g }, backend: n, attrs: { shape: l } });
    f[m] = b, p.push(g);
  }
  return p.forEach((m) => n.disposeData(m.dataId)), f;
}
var $oe = { kernelName: Ko, backendName: "webgpu", kernelFunc: Toe };
var _oe = [Zne, Pse, Mse, Vse, jse, Xse, Qse, Jse, rre, ure, cre, fre, nse, yre, Sre, Tre, _re, Ere, Fre, Pre, Mre, Vre, Ure, Kre, Yre, Zre, Jre, eae, nae, rae, iae, pae, uae, cae, mae, bae, vae, kae, Cae, Tae, _ae, tse, gre, Eae, Dae, Oae, zae, Lae, Vae, Wae, Gae, qae, Kae, Yae, Zae, eie, Gre, nie, rie, iie, are, uie, cie, pie, fie, gie, yie, xie, ire, wie, Sie, Cie, Yne, $ie, Eie, Die, Oie, Mie, Vie, Uie, Hie, jie, nre, doe, hoe, Yie, Zie, toe, soe, aoe, ioe, uoe, Kie, qre, moe, voe, Soe, Noe, Hse, $oe, oie];
for (let e of _oe)
  Fl(e);
var Aoe = class {
  constructor(e) {
    this.device = e, this.numUsedBuffers = 0, this.numFreeBuffers = 0, this.freeBuffers = /* @__PURE__ */ new Map(), this.usedBuffers = /* @__PURE__ */ new Map(), this.numBytesUsed = 0, this.numBytesAllocated = 0;
  }
  acquireUploadBuffer(e, t) {
    return this.acquireBuffer(e, t, true);
  }
  acquireBuffer(e, t, n = false) {
    let s = Uw(e, t);
    if (this.freeBuffers.has(s) || this.freeBuffers.set(s, []), this.usedBuffers.has(s) || this.usedBuffers.set(s, []), this.numBytesUsed += e, this.numUsedBuffers++, this.freeBuffers.get(s).length > 0) {
      this.numFreeBuffers--;
      let a = this.freeBuffers.get(s).shift();
      return this.usedBuffers.get(s).push(a), a;
    }
    this.numBytesAllocated += e;
    let r = this.device.createBuffer({ mappedAtCreation: n, size: e, usage: t });
    return this.usedBuffers.get(s).push(r), r;
  }
  releaseBuffer(e, t, n) {
    if (this.freeBuffers.size === 0)
      return;
    let s = Uw(t, n);
    this.freeBuffers.has(s) || this.freeBuffers.set(s, []), this.freeBuffers.get(s).push(e), this.numFreeBuffers++, this.numUsedBuffers--;
    let r = this.usedBuffers.get(s), a = r.indexOf(e);
    if (a < 0)
      throw new Error("Cannot release a buffer that was never provided by this buffer manager");
    r.splice(a, 1), this.numBytesUsed -= t;
  }
  releaseUploadBuffer(e, t, n) {
    e.mapAsync(GPUMapMode.WRITE).then(() => {
      this.releaseBuffer(e, t, n);
    }, (s) => {
    });
  }
  getNumUsedBuffers() {
    return this.numUsedBuffers;
  }
  getNumFreeBuffers() {
    return this.numFreeBuffers;
  }
  dispose() {
    this.freeBuffers.forEach((e, t) => {
      e.forEach((n) => {
        n.destroy();
      });
    }), this.usedBuffers.forEach((e, t) => {
      e.forEach((n) => {
        n.destroy();
      });
    }), this.freeBuffers = /* @__PURE__ */ new Map(), this.usedBuffers = /* @__PURE__ */ new Map(), this.numUsedBuffers = 0, this.numFreeBuffers = 0, this.numBytesUsed = 0, this.numBytesAllocated = 0;
  }
};
function Uw(e, t) {
  return `${e}_${t}`;
}
var Eoe = class {
  constructor(e) {
    this.device = e, this.numUsedTextures = 0, this.numFreeTextures = 0, this.freeTextures = /* @__PURE__ */ new Map(), this.usedTextures = /* @__PURE__ */ new Map(), this.numBytesUsed = 0, this.numBytesAllocated = 0;
  }
  acquireTexture(e, t, n, s) {
    let r = Hw(n), a = e * t * r, i = Gw(e, t, n, s);
    if (this.freeTextures.has(i) || this.freeTextures.set(i, []), this.usedTextures.has(i) || this.usedTextures.set(i, []), this.numBytesUsed += a, this.numUsedTextures++, this.freeTextures.get(i).length > 0) {
      this.numFreeTextures--;
      let u = this.freeTextures.get(i).shift();
      return this.usedTextures.get(i).push(u), u;
    }
    this.numBytesAllocated += a;
    let o = this.device.createTexture({ size: [e, t], format: n, usage: s });
    return this.usedTextures.get(i).push(o), o;
  }
  releaseTexture(e, t, n, s, r) {
    if (this.freeTextures.size === 0)
      return;
    let a = Gw(t, n, s, r);
    this.freeTextures.has(a) || this.freeTextures.set(a, []), this.freeTextures.get(a).push(e), this.numFreeTextures++, this.numUsedTextures--;
    let i = this.usedTextures.get(a), o = i.indexOf(e);
    if (o < 0)
      throw new Error("Cannot release a texture that was never provided by this texture manager");
    i.splice(o, 1);
    let u = Hw(s), l = t * n * u;
    this.numBytesUsed -= l;
  }
  getNumUsedTextures() {
    return this.numUsedTextures;
  }
  getNumFreeTextures() {
    return this.numFreeTextures;
  }
  dispose() {
    this.freeTextures.forEach((e, t) => {
      e.forEach((n) => {
        n.destroy();
      });
    }), this.usedTextures.forEach((e, t) => {
      e.forEach((n) => {
        n.destroy();
      });
    }), this.freeTextures = /* @__PURE__ */ new Map(), this.usedTextures = /* @__PURE__ */ new Map(), this.numUsedTextures = 0, this.numFreeTextures = 0, this.numBytesUsed = 0, this.numBytesAllocated = 0;
  }
};
function Gw(e, t, n, s) {
  return `${e}_${t}_${n}_${s}`;
}
function Hw(e) {
  if (e === "rgba8unorm")
    return 16;
  throw new Error(`${e} is not supported!`);
}
var Roe = (e, t, n, s, r) => {
  let a = [s, ...n];
  return r && a.push(r), e.createBindGroup({ layout: t, entries: a.map((i, o) => ({ binding: o, resource: i })) });
};
var qw = (e, t, n, s, r, a = false) => {
  let i = { dtype: r.dtype, shape: r.shape }, o = Pne(s, i, t, a), u = e.createShaderModule({ code: o, label: t.constructor.name });
  return e.createComputePipeline({ layout: n, compute: { module: u, entryPoint: "main" }, label: t.constructor.name });
};
function jw(e, t, n = [], s = "", r = "") {
  return e.shaderKey + "_" + (e.workGroupSize ? e.workGroupSize.join(",") : "") + t.map((i) => i.length).join(",") + n.join(",") + e.variableNames.join(",") + s + r;
}
var Doe = K().getNumber("WEBGPU_CPU_HANDOFF_SIZE_THRESHOLD");
var Kw = (e, t) => {
  let n = e.limits.maxComputeWorkgroupsPerDimension, s = t.dispatchLayout, r = t.dispatch;
  if (r.every((i) => i <= n))
    return r;
  w.assert(r[0] > n && s.y === void 0 && s.z === void 0, () => "Dispatch size exceeds WebGPU limits in Y or Z dimension.");
  let a = Math.ceil(Math.sqrt(r[0]));
  return a > n ? (a = Math.ceil(Math.cbrt(r[0])), w.assert(a <= n, () => "Total dispatch size exceeds WebGPU maximum."), [a, a, a]) : [a, a, 1];
};
var K2 = class extends il {
  constructor(e, t = false) {
    if (super(), this.commandQueueOwnedIds = /* @__PURE__ */ new WeakSet(), this.tensorDisposalQueue = [], this.uniformDisposalQueue = [], this.stagingDisposalQueue = [], this.textureDisposalQueue = [], this.disposed = false, this.uploadWaitMs = 0, this.downloadWaitMs = 0, this.dispatchNumberInEncoder = 0, this.fromPixelTextureLayout = null, this.fromPixelImportTextureLayout = null, !Fv())
      throw new Error("WebGPU is not supported on this device");
    this.layoutCache = {}, this.pipelineCache = {}, this.device = e, this.queue = e.queue, this.currentCommandEncoder = null, this.currentComputePass = null, this.supportTimeQuery = t, this.bufferManager = new Aoe(this.device), this.textureManager = new Eoe(this.device), this.tensorMap = new Yd(this, ds()), this.supportTimeQuery && (this.querySet = this.device.createQuerySet({ type: "timestamp", count: 2 })), K().getBool("WEBGPU_USE_PROFILE_TOOL") && (this.dummyCanvas = document.createElement("canvas"), this.dummyCanvas.width = 1, this.dummyCanvas.height = 1, this.dummyContext = this.dummyCanvas.getContext("webgpu"), this.dummyContext.configure({ device: e, format: "bgra8unorm" }), document.body.appendChild(this.dummyCanvas));
  }
  nextDataId() {
    return K2.nextDataId++;
  }
  floatPrecision() {
    return 32;
  }
  defaultGpuBufferUsage() {
    return GPUBufferUsage.STORAGE | GPUBufferUsage.COPY_SRC | GPUBufferUsage.COPY_DST;
  }
  flushDisposalQueue() {
    this.tensorDisposalQueue.forEach((e) => {
      this.maybeReleaseBuffer(e), this.tensorMap.delete(e);
    }), this.uniformDisposalQueue.forEach((e) => this.bufferManager.releaseBuffer(e.buffer, e.byteSize, e.usage)), this.stagingDisposalQueue.forEach((e) => this.bufferManager.releaseUploadBuffer(e.buffer, e.byteSize, e.usage)), this.textureDisposalQueue.forEach((e) => this.textureManager.releaseTexture(e.texture, e.width, e.height, e.format, e.usage)), this.tensorDisposalQueue = [], this.uniformDisposalQueue = [], this.stagingDisposalQueue = [], this.textureDisposalQueue = [];
  }
  disposeData(e, t = false) {
    if (this.tensorMap.has(e)) {
      let n = this.tensorMap.get(e);
      if (n.refCount--, !t && n.refCount > 0)
        return false;
      if (this.commandQueueOwnedIds.has(e))
        return this.tensorDisposalQueue.push(e), false;
      this.maybeReleaseBuffer(e);
      let { complexTensorInfos: s } = this.tensorMap.get(e);
      s != null && (this.disposeData(s.real.dataId, true), this.disposeData(s.imag.dataId, true)), this.tensorMap.delete(e);
    }
    return true;
  }
  memory() {
    return { numBytesInGPU: this.bufferManager.numBytesUsed, numBytesAllocatedInGPU: this.bufferManager.numBytesAllocated, unreliable: false };
  }
  getBufferManager() {
    return this.bufferManager;
  }
  getTextureManager() {
    return this.textureManager;
  }
  acquireBuffer(e, t = this.defaultGpuBufferUsage()) {
    return this.bufferManager.acquireBuffer(e, t);
  }
  maybeReleaseBuffer(e) {
    let t = this.tensorMap.get(e);
    t != null && t.bufferInfo.buffer != null && (this.bufferManager.releaseBuffer(t.bufferInfo.buffer, t.bufferInfo.byteSize, t.bufferInfo.usage), t.bufferInfo.buffer = null);
  }
  refCount(e) {
    return this.tensorMap.has(e) ? this.tensorMap.get(e).refCount : 0;
  }
  incRef(e) {
    let t = this.tensorMap.get(e);
    t.refCount++;
  }
  decRef(e) {
    if (this.tensorMap.has(e)) {
      let t = this.tensorMap.get(e);
      t.refCount--;
    }
  }
  write(e, t, n) {
    if (n === "complex64" && e != null)
      throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");
    let s = { id: this.nextDataId() }, r = w.sizeFromShape(t) * fd(n);
    return this.tensorMap.set(s, { dtype: n, shape: t, values: e, bufferInfo: { byteSize: r, usage: this.defaultGpuBufferUsage() }, refCount: 1 }), s;
  }
  move(e, t, n, s, r) {
    if (s === "complex64")
      throw new Error("Cannot write to a complex64 dtype. Please use tf.complex(real, imag).");
    let a = w.sizeFromShape(n) * fd(s);
    this.tensorMap.set(e, { dtype: s, shape: n, values: t, bufferInfo: { byteSize: a, usage: this.defaultGpuBufferUsage() }, refCount: r });
  }
  submitQueue() {
    this.ensureComputePassEnded(), this.queue.submit([this.currentCommandEncoder.finish()]), this.currentCommandEncoder = null, this.dispatchNumberInEncoder = 0, this.commandQueueOwnedIds = /* @__PURE__ */ new WeakSet(), this.flushDisposalQueue();
  }
  getBuffer(e) {
    return this.uploadToGPU(e), this.tensorMap.get(e).bufferInfo.buffer;
  }
  ensureCommandEncoderReady() {
    this.currentCommandEncoder || (this.currentCommandEncoder = this.device.createCommandEncoder());
  }
  ensureComputePassEnded() {
    this.currentComputePass && (this.currentComputePass.end(), this.currentComputePass = null);
  }
  getComputePass() {
    return this.currentComputePass || (this.currentComputePass = this.currentCommandEncoder.beginComputePass()), this.currentComputePass;
  }
  async getBufferData(e, t) {
    let n = this.acquireBuffer(t, GPUBufferUsage.COPY_DST | GPUBufferUsage.MAP_READ);
    this.ensureCommandEncoderReady(), this.ensureComputePassEnded(), this.currentCommandEncoder.copyBufferToBuffer(e, 0, n, 0, t), this.submitQueue(), await n.mapAsync(GPUMapMode.READ);
    let s = n.getMappedRange().slice(0);
    return n.unmap(), n != null && this.bufferManager.releaseBuffer(n, t, GPUBufferUsage.COPY_DST | GPUBufferUsage.MAP_READ), K().getBool("WEBGPU_USE_PROFILE_TOOL") && (w.assert(this.dummyContext !== void 0, () => "Fail to get context for profiling tool"), this.dummyContext.getCurrentTexture()), s;
  }
  convertAndCacheOnCPU(e, t) {
    let n = this.tensorMap.get(e);
    return this.maybeReleaseBuffer(e), n.values = t, n.values;
  }
  readSync(e) {
    let t = this.tensorMap.get(e), { values: n } = t;
    if (n == null)
      throw new Error("WebGPU readSync is only available for CPU-resident tensors.");
    return n;
  }
  async read(e) {
    if (!this.tensorMap.has(e))
      throw new Error(`Tensor ${e} was not registered!`);
    let t = this.tensorMap.get(e), { values: n } = t;
    if (n != null)
      return this.convertAndCacheOnCPU(e, n);
    let s;
    if (t.dtype === "complex64") {
      let r = await Promise.all([this.read(t.complexTensorInfos.real.dataId), this.read(t.complexTensorInfos.imag.dataId)]), a = r[0], i = r[1];
      s = C.mergeRealAndImagArrays(a, i);
    } else {
      let r = t.values != null ? t.values : await this.getBufferData(t.bufferInfo.buffer, t.bufferInfo.byteSize);
      s = $2(r, t.dtype);
    }
    return this.convertAndCacheOnCPU(e, s), s;
  }
  readToGPU(e) {
    let t = this.tensorMap.get(e), { values: n, dtype: s, shape: r, bufferInfo: a } = t;
    if (s === "complex64")
      throw new Error("Does not support reading buffer for complex64 dtype.");
    if (a.buffer == null)
      throw n != null ? new Error("Data is not on GPU but on CPU.") : new Error("There is no data on GPU or CPU.");
    let i = w.sizeFromShape(r) * fd(s), o = this.acquireBuffer(i);
    this.ensureCommandEncoderReady(), this.ensureComputePassEnded(), this.currentCommandEncoder.copyBufferToBuffer(a.buffer, 0, o, 0, i), this.submitQueue();
    let u = this.makeTensorInfo(r, s), l = ds().makeTensorFromTensorInfo(u), c = this.tensorMap.get(u.dataId);
    return c.bufferInfo.buffer = o, { tensorRef: l, buffer: o, bufSize: i };
  }
  bufferSync(e) {
    let t = this.readSync(e.dataId);
    if (e.dtype === "string")
      try {
        let n = t.map((s) => w.decodeString(s));
        return Ae(e.shape, e.dtype, n);
      } catch (n) {
        throw new Error("Failed to decode encoded string bytes into utf-8");
      }
    return Ae(e.shape, e.dtype, t);
  }
  async time(e) {
    let t = this.activeTimers, n = [], s = false;
    this.programTimersStack == null ? (this.programTimersStack = n, s = true) : this.activeTimers.push(n), this.activeTimers = n, e();
    let r = w.flatten(this.activeTimers.map((u) => u.query)).filter((u) => u != null), a = w.flatten(this.activeTimers.map((u) => u.name)).filter((u) => u != null);
    this.activeTimers = t, s && (this.programTimersStack = null);
    let i = { uploadWaitMs: this.uploadWaitMs, downloadWaitMs: this.downloadWaitMs, kernelMs: null, wallMs: null }, o = await Promise.all(r);
    return i.kernelMs = w.sum(o), i.getExtraProfileInfo = () => o.map((u, l) => ({ name: a[l], ms: u })).map((u) => `${u.name}: ${u.ms}`).join(", "), this.uploadWaitMs = 0, this.downloadWaitMs = 0, i;
  }
  getAndSavePipeline(e, t) {
    return e in this.pipelineCache || (this.pipelineCache[e] = t()), this.pipelineCache[e];
  }
  makeTensorInfo(e, t, n) {
    let s;
    if (t === "string" && n != null && n.length > 0 && w.isString(n[0])) {
      let r = n.map((a) => w.encodeString(a));
      s = this.write(r, e, t);
    } else
      s = this.write(n, e, t);
    return { dataId: s, shape: e, dtype: t };
  }
  tensorToBinding(e) {
    if (!e)
      return null;
    let t = this.tensorMap.get(e.dataId);
    return { offset: 0, size: t.bufferInfo.byteSize, buffer: t.bufferInfo.buffer };
  }
  async getQueryTime(e) {
    return this.supportTimeQuery ? this.getTimeFromQuerySet(e) : 0;
  }
  uploadToGPU(e) {
    let t = this.tensorMap.get(e);
    if (t.bufferInfo.buffer == null && (t.bufferInfo.buffer = this.acquireBuffer(t.bufferInfo.byteSize), t.values)) {
      let n = this.bufferManager.acquireUploadBuffer(t.bufferInfo.byteSize, GPUBufferUsage.MAP_WRITE | GPUBufferUsage.COPY_SRC), s = n.getMappedRange();
      t.dtype === "int32" || t.dtype === "bool" ? new Int32Array(s).set(t.values) : new Float32Array(s).set(t.values), n.unmap(), this.ensureCommandEncoderReady(), this.ensureComputePassEnded(), this.currentCommandEncoder.copyBufferToBuffer(n, 0, t.bufferInfo.buffer, 0, t.bufferInfo.byteSize);
      let r = { byteSize: t.bufferInfo.byteSize, usage: GPUBufferUsage.MAP_WRITE | GPUBufferUsage.COPY_SRC, buffer: n };
      this.stagingDisposalQueue.push(r);
    }
  }
  makeUniforms(e) {
    let t = 0, n = 0, s = [];
    e.forEach((o) => {
      o.data.length === 0 && (o.data = [1]);
      let u;
      switch (o.data.length) {
        case 1:
          u = 4;
          break;
        case 2:
          u = 8;
          break;
        case 3:
          u = 16;
          break;
        case 4:
          u = 16;
          break;
        case 5:
          u = 16;
          break;
        case 6:
          u = 16;
          break;
        default:
          w.assert(false, () => `Unsupported ${o.data.length}D shape`);
      }
      (n === 5 || n === 6) && (u = 16), t = Math.ceil(t / u) * u, n = o.data.length, s.push(t), t += o.data.length * 4;
    });
    let r = new ArrayBuffer(t);
    e.forEach((o, u) => {
      let l = s[u];
      o.type === "int32" ? new Int32Array(r, l, o.data.length).set(o.data) : o.type === "uint32" ? new Uint32Array(r, l, o.data.length).set(o.data) : new Float32Array(r, l, o.data.length).set(o.data);
    });
    let a = this.acquireBuffer(t, GPUBufferUsage.COPY_DST | GPUBufferUsage.UNIFORM);
    this.queue.writeBuffer(a, 0, r, 0, t);
    let i = { byteSize: t, usage: GPUBufferUsage.COPY_DST | GPUBufferUsage.UNIFORM, buffer: a };
    return this.uniformDisposalQueue.push(i), { offset: 0, size: t, buffer: a };
  }
  createLayout(e) {
    let t = [];
    t.push({ binding: 0, visibility: GPUShaderStage.COMPUTE, buffer: { type: "storage" } });
    for (let r = 0; r < e; r++)
      t.push({ binding: r + 1, visibility: GPUShaderStage.COMPUTE, buffer: { type: "read-only-storage" } });
    t.push({ binding: e + 1, visibility: GPUShaderStage.COMPUTE, buffer: { type: "uniform" } });
    let n = this.device.createBindGroupLayout({ entries: t }), s = this.device.createPipelineLayout({ bindGroupLayouts: [n] });
    return { bindGroupLayout: n, pipelineLayout: s };
  }
  getCachedOrCreateLayout(e) {
    return e in this.layoutCache || (this.layoutCache[e] = this.createLayout(e)), this.layoutCache[e];
  }
  runWebGPUProgram(e, t, n, s, r) {
    if (!r) {
      if (r = this.makeTensorInfo(e.outputShape, n), w.sizeFromShape(r.shape) === 0) {
        let I = this.tensorMap.get(r.dataId);
        return I.values = w.getTypedArrayFromDType(r.dtype, 0), r;
      }
      this.uploadToGPU(r.dataId);
    }
    e.dispatch = Kw(this.device, e);
    let a = [{ type: "float32", data: [NaN] }], i = t.concat(r).map((I) => I.shape), o = "int32";
    i.map((I) => {
      a.push({ type: o, data: I });
    });
    let u = w.computeStrides(r.shape);
    if (a.push({ type: o, data: u }), e.size) {
      let I = w.sizeFromShape(e.outputShape);
      a.push({ type: o, data: [e.isVec4 ? I / 4 : I] });
    }
    s && (a = [...a, ...s]);
    let l = this.makeUniforms(a), c = t.map((I, $) => {
      if (I.dtype === "complex64")
        throw new Error("GPGPUProgram does not support complex64 input. For complex64 dtypes, please separate the program into real and imaginary parts.");
      return this.uploadToGPU(I.dataId), { dtype: this.tensorMap.get(I.dataId).dtype, shape: I.shape, name: e.variableNames[$] };
    }), p = c.map((I) => I.dtype).concat(r.dtype), d = c.map((I) => C.getBroadcastDims(I.shape, r.shape)), h = c.map((I) => w.arraysEqual(I.shape, r.shape)).join("_"), f = d.map((I) => I.join("_")).join(";"), m = jw(e, i, p, f, h), { bindGroupLayout: g, pipelineLayout: b } = this.getCachedOrCreateLayout(e.variableNames.length), y = this.getAndSavePipeline(m, () => qw(this.device, e, b, c, r)), v = this.activeTimers != null, x = Roe(this.device, g, t.map((I) => this.tensorToBinding(I)), this.tensorToBinding(r), l);
    this.ensureCommandEncoderReady();
    let k = this.getComputePass();
    return v && this.supportTimeQuery && k.writeTimestamp(this.querySet, 0), k.setPipeline(y), k.setBindGroup(0, x), k.dispatch(e.dispatch[0], e.dispatch[1], e.dispatch[2]), v && this.supportTimeQuery && k.writeTimestamp(this.querySet, 1), this.dispatchNumberInEncoder++, t.forEach((I) => {
      this.commandQueueOwnedIds.add(I.dataId);
    }), this.commandQueueOwnedIds.add(r.dataId), K().get("WEBGPU_DEFERRED_SUBMIT_BATCH_SIZE") <= this.dispatchNumberInEncoder && this.submitQueue(), v && this.activeTimers.push({ name: e.constructor.name, query: this.getQueryTime(this.querySet) }), r;
  }
  getFromPixelTextureLayout(e) {
    return e ? (this.fromPixelImportTextureLayout === null && (this.fromPixelImportTextureLayout = this.createFromPixelTextureLayout(true)), this.fromPixelImportTextureLayout) : (this.fromPixelTextureLayout === null && (this.fromPixelTextureLayout = this.createFromPixelTextureLayout(false)), this.fromPixelTextureLayout);
  }
  createFromPixelTextureLayout(e) {
    let t = [];
    t.push({ binding: 0, visibility: GPUShaderStage.COMPUTE, buffer: { type: "storage" } }), e ? t.push({ binding: 1, visibility: GPUShaderStage.COMPUTE, externalTexture: {} }) : t.push({ binding: 1, visibility: GPUShaderStage.COMPUTE, texture: {} }), t.push({ binding: 2, visibility: GPUShaderStage.COMPUTE, buffer: {} });
    let n = this.device.createBindGroupLayout({ entries: t }), s = this.device.createPipelineLayout({ bindGroupLayouts: [n] });
    return { bindGroupLayout: n, pipelineLayout: s };
  }
  copyExternalImageToTexture(e, t) {
    let n = GPUTextureUsage.COPY_DST | GPUTextureUsage.RENDER_ATTACHMENT | GPUTextureUsage.TEXTURE_BINDING, s = "rgba8unorm", r = this.textureManager.acquireTexture(t[1], t[0], s, n), a = r.createView();
    this.queue.copyExternalImageToTexture({ source: e }, { texture: r }, [t[1], t[0]]);
    let i = { width: t[1], height: t[0], format: s, usage: n, texture: r };
    return this.textureDisposalQueue.push(i), a;
  }
  runFromPixelsProgram(e, t, n, s, r) {
    e.dispatch = Kw(this.device, e);
    let a = this.makeTensorInfo(t, "int32");
    if (w.sizeFromShape(a.shape) === 0) {
      let m = this.tensorMap.get(a.dataId);
      return m.values = w.getTypedArrayFromDType(a.dtype, 0), a;
    }
    this.uploadToGPU(a.dataId);
    let i = jw(e, [a.shape]), o = this.getFromPixelTextureLayout(s), u = this.getAndSavePipeline(i, () => qw(this.device, e, o.pipelineLayout, [], a, true)), l;
    if (s) {
      let m = { source: r };
      l = this.device.importExternalTexture(m);
    } else
      l = this.copyExternalImageToTexture(r, a.shape);
    let c = this.tensorToBinding(a), p = this.makeUniforms(n), d = this.device.createBindGroup({ layout: o.bindGroupLayout, entries: [{ binding: 0, resource: { buffer: c.buffer } }, { binding: 1, resource: l }, { binding: 2, resource: { buffer: p.buffer } }] });
    this.ensureCommandEncoderReady();
    let h = this.getComputePass(), f = this.activeTimers != null;
    return f && this.supportTimeQuery && h.writeTimestamp(this.querySet, 0), h.setPipeline(u), h.setBindGroup(0, d), h.dispatch(e.dispatch[0], e.dispatch[1], e.dispatch[2]), f && this.supportTimeQuery && h.writeTimestamp(this.querySet, 1), this.commandQueueOwnedIds.add(a.dataId), this.dispatchNumberInEncoder++, K().get("WEBGPU_DEFERRED_SUBMIT_BATCH_SIZE") <= this.dispatchNumberInEncoder && this.submitQueue(), f && this.activeTimers.push({ name: e.constructor.name, query: this.getQueryTime(this.querySet) }), a;
  }
  async getTimeFromQuerySet(e) {
    let t = this.acquireBuffer(16, GPUBufferUsage.COPY_SRC | GPUBufferUsage.QUERY_RESOLVE), n = this.acquireBuffer(16, GPUBufferUsage.MAP_READ | GPUBufferUsage.COPY_DST);
    this.ensureCommandEncoderReady(), this.ensureComputePassEnded(), this.currentCommandEncoder.resolveQuerySet(e, 0, 2, t, 0), this.currentCommandEncoder.copyBufferToBuffer(t, 0, n, 0, 16), this.submitQueue(), await n.mapAsync(GPUMapMode.READ);
    let s = new BigUint64Array(n.getMappedRange()), r = Number(s[1] - s[0]);
    return n.unmap(), this.bufferManager.releaseBuffer(n, 16, GPUBufferUsage.MAP_READ | GPUBufferUsage.COPY_DST), this.bufferManager.releaseBuffer(t, 16, GPUBufferUsage.COPY_SRC | GPUBufferUsage.QUERY_RESOLVE), r / 1e6;
  }
  shouldExecuteOnCPU(e, t = Doe) {
    return K().getBool("WEBGPU_CPU_FORWARD") && e.every((n) => this.tensorMap.get(n.dataId).bufferInfo.buffer == null && w.sizeFromShape(n.shape) < t);
  }
  numDataIds() {
    return this.tensorMap.numDataIds() - this.tensorDisposalQueue.length;
  }
  dispose() {
    this.disposed || (this.bufferManager.dispose(), this.textureManager.dispose(), this.disposed = true);
  }
};
var Mv = K2;
Mv.nextDataId = 0;
var Foe = {};
Ee(Foe, { WebGPUBackend: () => Mv, webgpu_util: () => N2 });
Fv() && vp("webgpu", async () => {
  K().set("CHECK_COMPUTATION_FOR_ERRORS", false);
  let e = { powerPreference: K().get("WEBGPU_USE_LOW_POWER_GPU") ? "low-power" : "high-performance" }, t = await navigator.gpu.requestAdapter(e), n = t.limits, s = {}, r = t.features.has("timestamp-query");
  s.requiredLimits = { maxComputeWorkgroupStorageSize: n.maxComputeWorkgroupStorageSize, maxComputeWorkgroupsPerDimension: n.maxComputeWorkgroupsPerDimension }, r ? s.requiredFeatures = ["timestamp-query"] : console.warn("This device doesn't support timestamp-query extension. Start Chrome browser with flag --disable-dawn-features=disallow_unsafe_apis then try again. Or zero will shown for the kernel time when profiling mode isenabled. Using performance.now is not workable for webgpu sinceit doesn't support synchronously to read data from GPU.");
  let a = await t.requestDevice(s);
  return new Mv(a, r);
}, 3);
var St = ((e) => (e[e.float32 = 0] = "float32", e[e.int32 = 1] = "int32", e[e.bool = 2] = "bool", e[e.string = 3] = "string", e[e.complex64 = 4] = "complex64", e))(St || {});
var ih = ((e) => (e[e.linear = 0] = "linear", e[e.relu = 1] = "relu", e[e.relu6 = 2] = "relu6", e[e.prelu = 3] = "prelu", e[e.leakyrelu = 4] = "leakyrelu", e[e.sigmoid = 5] = "sigmoid", e[e.elu = 6] = "elu", e))(ih || {});
var X2;
function Ooe(e) {
  X2 = e.wasm.cwrap(aa, null, ["number", "array", "number", "number", "array", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function Poe(e) {
  let { inputs: t, backend: n, attrs: s } = e, { a: r, b: a, bias: i, preluActivationWeights: o } = t;
  if (r.dtype !== "float32" || a.dtype !== "float32")
    throw new Error("_FusedMatMul for non non-float32 tensors not yet supported.");
  let { transposeA: u, transposeB: l, activation: c, leakyreluAlpha: p } = s, d = n.dataIdMap.get(r.dataId).id, h = n.dataIdMap.get(a.dataId).id, f = 0;
  if (i != null) {
    let R = n.dataIdMap.get(i.dataId);
    if (R.shape.length !== 1)
      throw new Error(`_FusedMatMul only supports rank-1 bias but got rank ${R.shape.length}.`);
    f = R.id;
  }
  let m = o == null ? 0 : n.dataIdMap.get(o.dataId).id, g = ih[c];
  if (g == null)
    throw new Error(`${c} activation not yet supported for FusedConv2D in the wasm backend.`);
  let b = u ? r.shape[2] : r.shape[1], y = l ? a.shape[1] : a.shape[2], v = Qo.assertAndGetBroadcastShape(r.shape.slice(0, -2), a.shape.slice(0, -2)), x = n.makeOutput([...v, b, y], r.dtype), k = n.dataIdMap.get(x.dataId).id, I = new Uint8Array(new Int32Array(r.shape).buffer), $ = new Uint8Array(new Int32Array(a.shape).buffer);
  return X2(d, I, r.shape.length, h, $, a.shape.length, u, l, g, f, m, p || 0, k), x;
}
var zoe = { kernelName: aa, backendName: "wasm", setupFunc: Ooe, kernelFunc: Poe };
function Xt(e, t) {
  let n;
  function s(a) {
    n = a.wasm.cwrap(e, null, ["number", "number", "number"]);
  }
  function r(a) {
    let { backend: i, inputs: { x: o } } = a, u = i.dataIdMap.get(o.dataId).id, l = i.makeOutput(o.shape, t || o.dtype), c = i.dataIdMap.get(l.dataId).id;
    return w.sizeFromShape(l.shape) === 0 || n(u, St[o.dtype], c), l;
  }
  return { kernelName: e, backendName: "wasm", setupFunc: s, kernelFunc: r };
}
var Moe = Xt(po);
function gn(e, t, n) {
  let s;
  function r(i) {
    s = i.wasm.cwrap(e, null, ["number", "array", "number", "number", "array", "number", "number", "number"]);
  }
  function a(i) {
    let { backend: o, inputs: u } = i, { a: l, b: c } = u, p = o.dataIdMap.get(l.dataId).id, d = o.dataIdMap.get(c.dataId).id, h = n != null ? n : l.dtype, f = C.assertAndGetBroadcastShape(l.shape, c.shape), m = o.makeOutput(f, h);
    if (w.sizeFromShape(f) === 0)
      return m;
    let g = new Uint8Array(new Int32Array(l.shape).buffer), b = new Uint8Array(new Int32Array(c.shape).buffer), y = o.dataIdMap.get(m.dataId).id;
    return (() => s(p, g, l.shape.length, d, b, c.shape.length, St[l.dtype], y))(), m;
  }
  return { kernelName: e, backendName: "wasm", setupFunc: r, kernelFunc: a };
}
var Loe = true;
var Boe = gn(Sr, Loe);
var Y2;
function Voe(e) {
  Y2 = e.wasm.cwrap(Sa, null, ["array", "number", "number", "number"]);
}
function Woe(e) {
  let { inputs: t, backend: n } = e, s = n.makeOutput(t[0].shape, t[0].dtype);
  if (w.sizeFromShape(s.shape) === 0)
    return s;
  let r = t.map((o) => n.dataIdMap.get(o.dataId).id), a = new Uint8Array(new Int32Array(r).buffer), i = n.dataIdMap.get(s.dataId).id;
  return Y2(a, r.length, St[s.dtype], i), s;
}
var Uoe = { kernelName: Sa, backendName: "wasm", setupFunc: Voe, kernelFunc: Woe };
function oh(e) {
  let { inputs: { x: t }, backend: n } = e, s = n.makeOutput(t.shape, t.dtype), r = n.typedArrayFromHeap(t);
  return n.typedArrayFromHeap(s).set(r), s;
}
var Goe = { kernelName: Wa, backendName: "wasm", kernelFunc: oh };
var Q2;
function Hoe(e) {
  Q2 = e.wasm.cwrap(mi, null, ["number", "array", "number", "number", "number", "array", "number"]);
}
function wr(e) {
  let { inputs: t, backend: n, attrs: s } = e, [r, a] = joe(t.x.shape, s.perm), i = true;
  for (let f = 0; f < a.length; f++)
    a[f] !== f && (i = false);
  let o = qoe(t.x.shape, s.perm), u = { dataId: t.x.dataId, shape: r, dtype: t.x.dtype };
  if (i) {
    let f = oh({ inputs: t, backend: n });
    return f.shape = o, f;
  }
  let l = n.makeOutput(o, u.dtype), c = n.dataIdMap.get(u.dataId).id, p = n.dataIdMap.get(l.dataId).id, d = new Uint8Array(new Int32Array(a).buffer), h = new Uint8Array(new Int32Array(u.shape).buffer);
  return Q2(c, h, u.shape.length, St[u.dtype], p, d, a.length), l;
}
function qoe(e, t) {
  let n = new Array(e.length);
  for (let s = 0; s < n.length; s++)
    n[s] = e[t[s]];
  return n;
}
function joe(e, t) {
  let n = [], s = [];
  for (let r = 0; r < e.length; ++r)
    e[r] !== 1 && n.push(e[r]), e[t[r]] !== 1 && s.push(t[r]);
  for (let r = 0; r < s.length; ++r) {
    let a = -1;
    for (let i = 0; i < s.length; ++i)
      s[i] >= r && (a === -1 || s[a] > s[i]) && (a = i);
    s[a] = r;
  }
  return [n, s];
}
var Koe = { kernelName: mi, backendName: "wasm", kernelFunc: wr, setupFunc: Hoe };
function Or(e, t, n) {
  let s = e.shape, r = e.shape.length, a = w.parseAxisParam(t, s), i = a, o = C.getAxesPermutation(i, r), u = null, l = false;
  if (o != null) {
    let c = new Array(r);
    for (let h = 0; h < c.length; h++)
      c[h] = s[o[h]];
    i = C.getInnerMostAxes(i.length, r), u = wr({ inputs: { x: e }, attrs: { perm: o }, backend: n });
    let p = n.dataIdMap.get(e.dataId).id;
    n.dataIdMap.get(u.dataId).id !== p && (l = true);
  }
  return { transposed: u, originalAxes: a, axes: i, inputWasTransposed: l };
}
var Z2;
function Xoe(e) {
  Z2 = e.wasm.cwrap(ll, null, ["number, number, number"]);
}
function Yoe(e) {
  let { backend: t, inputs: n, attrs: s } = e, { axis: r, keepDims: a } = s, { x: i } = n, u = t.dataIdMap.get(i.dataId).id, l = i, { transposed: c, axes: p, originalAxes: d, inputWasTransposed: h } = Or(i, r, t);
  if (h) {
    let v = t.dataIdMap.get(c.dataId).id;
    l = c, u = v;
  }
  let f = l.shape.length;
  C.assertAxesAreInnerMostDims("all", p, f);
  let [m, g] = C.computeOutAndReduceShapes(l.shape, p), b = w.sizeFromShape(g), y = t.makeOutput(m, i.dtype);
  if (w.sizeFromShape(l.shape) !== 0) {
    let v = t.dataIdMap.get(y.dataId).id;
    Z2(u, b, v);
  }
  if (h && t.disposeData(c.dataId), a) {
    let v = C.expandShapeToKeepDim(y.shape, d);
    y.shape = v;
  }
  return y;
}
var Qoe = { kernelName: ll, backendName: "wasm", setupFunc: Xoe, kernelFunc: Yoe };
var J2;
function Zoe(e) {
  J2 = e.wasm.cwrap(cl, null, ["number, number, number"]);
}
function Joe(e) {
  let { backend: t, inputs: n, attrs: s } = e, { axis: r, keepDims: a } = s, { x: i } = n, u = t.dataIdMap.get(i.dataId).id, l = i, { transposed: c, axes: p, originalAxes: d, inputWasTransposed: h } = Or(i, r, t);
  if (h) {
    let v = t.dataIdMap.get(c.dataId).id;
    l = c, u = v;
  }
  let f = l.shape.length;
  C.assertAxesAreInnerMostDims("any", p, f);
  let [m, g] = C.computeOutAndReduceShapes(l.shape, p), b = w.sizeFromShape(g), y = t.makeOutput(m, i.dtype);
  if (w.sizeFromShape(l.shape) !== 0) {
    let v = t.dataIdMap.get(y.dataId).id;
    J2(u, b, v);
  }
  if (h && t.disposeData(c.dataId), a) {
    let v = C.expandShapeToKeepDim(y.shape, d);
    y.shape = v;
  }
  return y;
}
var eue = { kernelName: cl, backendName: "wasm", setupFunc: Zoe, kernelFunc: Joe };
var eN;
function tue(e) {
  eN = e.wasm.cwrap(Ia, null, ["number", "number", "number", "number", "number"]);
}
function nue(e) {
  let { backend: t, inputs: n, attrs: s } = e, { axis: r } = s, { x: a } = n, i = t.dataIdMap.get(a.dataId).id, o = i, u = a, { transposed: l, axes: c, inputWasTransposed: p } = Or(a, r, t);
  if (p) {
    let b = t.dataIdMap.get(l.dataId).id;
    b !== i && (u = l, o = b);
  }
  let d = u.shape.slice(0, -1), h = t.makeOutput(d, "int32"), f = t.dataIdMap.get(h.dataId).id, m = w.sizeFromShape(h.shape), g = u.shape[c[0]];
  return eN(o, St[u.dtype], m, g, f), p && t.disposeData(l.dataId), h;
}
var sue = { kernelName: Ia, backendName: "wasm", kernelFunc: nue, setupFunc: tue };
var tN;
function rue(e) {
  tN = e.wasm.cwrap(Ca, null, ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function aue(e) {
  let { inputs: t, attrs: n, backend: s } = e, r = t.x, a = s.dataIdMap.get(r.dataId).id, { filterSize: i, strides: o, pad: u, dimRoundingMode: l } = n, c = C.computePool2DInfo(r.shape, i, o, 1, u, l), p = c.filterHeight, d = c.filterWidth, h = c.padInfo.top, f = c.padInfo.right, m = c.padInfo.bottom, g = c.padInfo.left, b = c.strideHeight, y = c.strideWidth, v = c.inChannels;
  if (c.dataFormat !== "channelsLast")
    throw new Error(`wasm backend does not support dataFormat:'${c.dataFormat}'. Please use 'channelsLast'.`);
  if (c.dilationWidth !== 1 || c.dilationHeight !== 1)
    throw new Error(`was backend only supports average pooling with dilation = [1, 1], got [${c.dilationHeight}, ${c.dilationWidth}].`);
  let x = s.makeOutput(c.outShape, "float32"), k = s.dataIdMap.get(x.dataId).id;
  return tN(a, r.shape[0], r.shape[1], r.shape[2], p, d, h, f, m, g, b, y, v, k), x;
}
var iue = { kernelName: Ca, backendName: "wasm", setupFunc: rue, kernelFunc: aue };
function yn(e) {
  let { inputs: t, attrs: n } = e, { x: s } = t, { shape: r } = n, a = w.sizeFromShape(s.shape), i = w.inferFromImplicitShape(r, a);
  return w.assert(a === w.sizeFromShape(i), () => `new shape: ${i}, old shape: ${s.shape}. New shape and old shape must have the same number of elements.`), e.backend.incRef(s.dataId), { dataId: s.dataId, shape: i, dtype: s.dtype };
}
var oue = { kernelName: Oo, backendName: "wasm", kernelFunc: yn };
var nN;
function uue(e) {
  nN = e.wasm.cwrap(Na, null, ["number", "array", "number", "number", "array", "number", "number", "number", "number"]);
}
function lue(e) {
  let { inputs: t, backend: n, attrs: s } = e, { a: r, b: a } = t, { transposeA: i, transposeB: o } = s;
  if (r.dtype !== "float32" || a.dtype !== "float32")
    throw new Error("BatchMatMul for non non-float32 tensors not yet supported.");
  let u = r.shape.length, l = a.shape.length, c = i ? r.shape[u - 2] : r.shape[u - 1], p = o ? a.shape[l - 1] : a.shape[l - 2], d = i ? r.shape[u - 1] : r.shape[u - 2], h = o ? a.shape[l - 2] : a.shape[l - 1], f = r.shape.slice(0, -2), m = a.shape.slice(0, -2), g = w.sizeFromShape(f), b = w.sizeFromShape(m), v = Qo.assertAndGetBroadcastShape(r.shape.slice(0, -2), a.shape.slice(0, -2)).concat([d, h]);
  w.assert(c === p, () => `Error in matMul: inner shapes (${c}) and (${p}) of Tensors with shapes ${r.shape} and ${a.shape} and transposeA=${i} and transposeB=${o} must match.`);
  let x = i ? [g, c, d] : [g, d, c], k = o ? [b, h, p] : [b, p, h], I = yn({ inputs: { x: r }, backend: n, attrs: { shape: x } }), $ = yn({ inputs: { x: a }, backend: n, attrs: { shape: k } }), R = n.dataIdMap.get(I.dataId).id, E = n.dataIdMap.get($.dataId).id, P = i ? I.shape[2] : I.shape[1], A = o ? $.shape[1] : $.shape[2], O = Math.max(g, b), T = n.makeOutput([O, P, A], I.dtype), z = n.dataIdMap.get(T.dataId).id, W = new Uint8Array(new Int32Array(I.shape).buffer), q = new Uint8Array(new Int32Array($.shape).buffer);
  return nN(R, W, I.shape.length, E, q, $.shape.length, i, o, z), n.disposeData(I.dataId), n.disposeData($.dataId), T.shape = v, T;
}
var cue = { kernelName: Na, backendName: "wasm", setupFunc: uue, kernelFunc: lue };
function xa(e) {
  let { inputs: { x: t }, attrs: { begin: n, size: s }, backend: r } = e, [a, i] = wt.parseSliceParams(t, n, s), o = wt.isSliceContinous(t.shape, a, i), u = r.readSync(t.dataId), l = r.makeOutput(i, t.dtype), c = w.computeStrides(t.shape), p = r.dataIdMap.get(l.dataId);
  if (o) {
    let f = wt.computeFlatOffset(a, c);
    return t.dtype === "string" ? p.stringBytes = u.slice(f, f + w.sizeFromShape(i)) : r.typedArrayFromHeap(l).set(u.subarray(f, f + w.sizeFromShape(i))), l;
  }
  if (t.dtype === "string") {
    let f = Bd(u, a, i, t.shape, t.dtype);
    return p.stringBytes = f, l;
  }
  let d = r.typedArrayFromHeap(l), h = t.shape.length;
  if (h === 2)
    due(u, c[0], d, a, i);
  else if (h === 3)
    pue(u, c[0], c[1], d, a, i);
  else if (h === 4)
    hue(u, c[0], c[1], c[2], d, a, i);
  else {
    let f = Bd(u, a, i, t.shape, t.dtype);
    d.set(f);
  }
  return l;
}
function due(e, t, n, s, r) {
  let a = 0, i = s[0], o = s[1], u = i + r[0];
  for (let l = i; l < u; l++) {
    let c = l * t + o;
    n.set(e.subarray(c, c + r[1]), a), a += r[1];
  }
}
function pue(e, t, n, s, r, a) {
  let i = 0, o = r[0], u = r[1], l = r[2], c = o + a[0], p = u + a[1];
  for (let d = o; d < c; d++)
    for (let h = u; h < p; h++) {
      let f = d * t + h * n + l;
      s.set(e.subarray(f, f + a[2]), i), i += a[2];
    }
}
function hue(e, t, n, s, r, a, i) {
  let o = 0, u = a[0], l = a[1], c = a[2], p = u + i[0], d = l + i[1], h = c + i[2], f = a[3];
  for (let m = u; m < p; m++)
    for (let g = l; g < d; g++)
      for (let b = c; b < h; b++) {
        let y = m * t + g * n + b * s + f;
        r.set(e.subarray(y, y + i[3]), o), o += i[3];
      }
}
var fue = { kernelName: Bo, backendName: "wasm", kernelFunc: xa };
function mue(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockShape: a, crops: i } = s, o = a.reduce((b, y) => b * y), u = C.getReshaped(r.shape, a, o), l = C.getPermuted(u.length, a.length), c = C.getReshapedPermuted(r.shape, a, o), p = C.getSliceBeginCoords(i, a.length), d = C.getSliceSize(c, i, a.length), h = yn({ inputs: { x: r }, backend: n, attrs: { shape: u } }), f = wr({ inputs: { x: h }, backend: n, attrs: { perm: l } }), m = yn({ inputs: { x: f }, backend: n, attrs: { shape: c } }), g = xa({ inputs: { x: m }, backend: n, attrs: { begin: p, size: d } });
  return n.disposeData(h.dataId), n.disposeData(f.dataId), n.disposeData(h.dataId), g;
}
var gue = { kernelName: ho, backendName: "wasm", kernelFunc: mue };
function oc(e) {
  let { inputs: { x: t }, attrs: { dtype: n }, backend: s } = e, r = s.makeOutput(t.shape, n), a = s.typedArrayFromHeap(t);
  return s.typedArrayFromHeap(r).set(a), r;
}
var bue = { kernelName: Ta, backendName: "wasm", kernelFunc: oc };
var yue = Xt($a);
var sN;
function vue(e) {
  sN = e.wasm.cwrap(Ir, null, ["number", "number", "number", "number"]);
}
function xue(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { clipValueMin: a, clipValueMax: i } = s, o = n.dataIdMap.get(r.dataId).id, u = n.makeOutput(r.shape, r.dtype), l = n.dataIdMap.get(u.dataId).id;
  return sN(o, a, i, l), u;
}
var wue = { kernelName: Ir, backendName: "wasm", setupFunc: vue, kernelFunc: xue };
function rN(e) {
  let { inputs: t, backend: n } = e, s = w.parseAxisParam(e.attrs.axis, t[0].shape)[0], r = C.computeOutShape(t.map((h) => h.shape), s), a = t.filter((h) => w.sizeFromShape(h.shape) > 0);
  if (a.length === 1)
    return oh({ inputs: { x: a[0] }, backend: n });
  let i = n.makeOutput(r, t[0].dtype);
  if (w.sizeFromShape(r) === 0)
    return i;
  let o = a.map((h) => h.shape);
  if (C.assertParamsConsistent(o, s), a[0].dtype === "string") {
    let h = a.map((v) => {
      let x = w.sizeFromShape(v.shape.slice(s));
      return yn({ inputs: { x: v }, backend: n, attrs: { shape: [-1, x] } });
    }), f = h.map((v) => ({ vals: n.readSync(v.dataId), shape: v.shape }));
    r = C.computeOutShape(h.map((v) => v.shape), 1);
    let m = h[0].shape[0] === 1, g = lv(f, r, t[0].dtype, m), b = C.computeOutShape(a.map((v) => v.shape), s);
    i.shape = b;
    let y = n.dataIdMap.get(i.dataId);
    return y.stringBytes = C.fromStringArrayToUint8(g), h.forEach((v) => n.disposeData(v.dataId)), i;
  }
  let u = w.sizeFromShape(a[0].shape.slice(0, s)), l = 0, c = a.map((h) => {
    let f = w.sizeFromShape(h.shape.slice(s));
    return l += f, f;
  }), p = a.map((h) => n.typedArrayFromHeap(h)), d = n.typedArrayFromHeap(i);
  for (let h = 0; h < u; h++) {
    let f = h * l;
    for (let m = 0; m < p.length; m++) {
      let g = c[m], b = h * g, y = p[m].subarray(b, b + g);
      d.set(y, f), f += g;
    }
  }
  return i;
}
var kue = { kernelName: fo, backendName: "wasm", kernelFunc: rN };
var aN;
function Sue(e) {
  aN = e.wasm.cwrap(_a, null, ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function Iue(e) {
  let { inputs: t, attrs: n, backend: s } = e, { x: r, filter: a } = t, i = s.dataIdMap.get(r.dataId).id, o = s.dataIdMap.get(a.dataId).id, { strides: u, dilations: l, pad: c, dimRoundingMode: p, dataFormat: d } = n, h = C.convertConv2DDataFormat(d), f = C.computeConv2DInfo(r.shape, a.shape, u, l, c, p, false, h), m = f.filterHeight, g = f.filterWidth, b = f.padInfo.top, y = f.padInfo.right, v = f.padInfo.bottom, x = f.padInfo.left, k = f.dilationHeight, I = f.dilationWidth, $ = f.strideHeight, R = f.strideWidth, E = f.inChannels, P = f.outChannels, A = f.padInfo.type === "SAME" ? 1 : 0;
  if (f.dataFormat !== "channelsLast")
    throw new Error(`wasm backend Conv2D does not support dataFormat:'${f.dataFormat}'. Please use 'channelsLast'.`);
  let O = s.makeOutput(f.outShape, "float32"), T = s.dataIdMap.get(O.dataId).id;
  return aN(i, r.shape[0], r.shape[1], r.shape[2], o, m, g, b, y, v, x, A, k, I, $, R, E, P, T), O;
}
var Cue = { kernelName: _a, backendName: "wasm", setupFunc: Sue, kernelFunc: Iue };
var iN;
function Nue(e) {
  iN = e.wasm.cwrap(Aa, null, ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function Tue(e) {
  let { backend: t, inputs: n, attrs: s } = e, { dy: r, filter: a } = n, { strides: i, pad: o, dataFormat: u, dimRoundingMode: l, inputShape: c } = s, p = 1, d = C.convertConv2DDataFormat(u), h = C.computeConv2DInfo(c, a.shape, i, p, o, l, false, d), { batchSize: f, filterHeight: m, filterWidth: g, inChannels: b, inHeight: y, inWidth: v, outChannels: x, outHeight: k, outWidth: I, strideHeight: $, strideWidth: R } = h, E = m - 1 - h.padInfo.top, P = g - 1 - h.padInfo.left, A = h.dataFormat === "channelsLast", O = w.computeStrides(h.inShape), T = w.computeStrides(r.shape), [z, W, q] = w.computeStrides(a.shape), X = O[0], Y = A ? O[1] : O[2], Z = A ? O[2] : 1, te = A ? 1 : O[1], J = T[0], se = A ? T[1] : T[2], ne = A ? T[2] : 1, oe = A ? 1 : T[1], ae = t.makeOutput(h.inShape, "float32"), de = t.dataIdMap.get(ae.dataId).id, me = t.dataIdMap.get(r.dataId).id, ke = t.dataIdMap.get(a.dataId).id;
  return iN(me, ke, f, m, g, y, v, b, k, I, x, $, R, E, P, z, W, q, X, Y, Z, te, J, se, ne, oe, de), ae;
}
var $ue = { kernelName: Aa, backendName: "wasm", setupFunc: Nue, kernelFunc: Tue };
var _ue = Xt(Ea);
var Aue = Xt(Ra);
var oN = ((e) => (e[e.bilinear = 0] = "bilinear", e[e.nearest = 1] = "nearest", e))(oN || {});
var uN;
function Eue(e) {
  uN = e.wasm.cwrap(go, null, ["number", "number", "number", "number", "array", "number", "number", "number", "number", "number"]);
}
function Rue(e) {
  let { backend: t, inputs: n, attrs: s } = e, { method: r, extrapolationValue: a, cropSize: i } = s, { image: o, boxes: u, boxInd: l } = n, c = u.shape[0], [p, d] = i, h = [c, p, d, o.shape[3]], f = t.dataIdMap.get(o.dataId), m;
  o.dtype !== "float32" && (m = oc({ backend: t, inputs: { x: o }, attrs: { dtype: "float32" } }), f = t.dataIdMap.get(m.dataId));
  let g = f.id, b = t.dataIdMap.get(u.dataId).id, y = t.dataIdMap.get(l.dataId).id, v = t.makeOutput(h, "float32"), x = t.dataIdMap.get(v.dataId).id, k = new Uint8Array(new Int32Array(o.shape).buffer);
  return uN(g, b, y, c, k, p, d, oN[r], a, x), m != null && t.disposeData(m.dataId), v;
}
var Due = { kernelName: go, backendName: "wasm", setupFunc: Eue, kernelFunc: Rue };
var lN;
function Fue(e) {
  lN = e.wasm.cwrap(mo, null, ["number", "number", "number", "number", "number", "number"]);
}
function Oue(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, exclusive: i, reverse: o } = s, u = r.shape.length;
  w.assert(r.dtype === "float32" || r.dtype === "int32", () => `cumprod does not support ${r.dtype} tensors in the WASM backend`);
  let l = C.getAxesPermutation([a], u), c = r;
  l !== null && (c = wr({ inputs: { x: r }, attrs: { perm: l }, backend: n }));
  let p = C.getInnerMostAxes(1, u)[0];
  C.assertAxesAreInnerMostDims("cumprod", [p], u);
  let d = n.makeOutput(c.shape, c.dtype), h = c.shape[p], f = n.dataIdMap.get(c.dataId).id, m = n.dataIdMap.get(d.dataId).id;
  lN(f, i ? 1 : 0, o ? 1 : 0, h, m, St[r.dtype]);
  let g = d;
  if (l !== null) {
    let b = C.getUndoAxesPermutation(l);
    g = wr({ inputs: { x: d }, attrs: { perm: b }, backend: n }), n.disposeData(c.dataId), n.disposeData(d.dataId);
  }
  return g;
}
var Pue = { kernelName: mo, backendName: "wasm", setupFunc: Fue, kernelFunc: Oue };
var cN;
function zue(e) {
  cN = e.wasm.cwrap(Da, null, ["number", "number", "number", "number", "number", "number"]);
}
function Mue(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { axis: a, exclusive: i, reverse: o } = s, u = r.shape.length;
  w.assert(r.dtype === "float32" || r.dtype === "int32", () => `cumsum does not support ${r.dtype} tensors in the WASM backend`);
  let l = C.getAxesPermutation([a], u), c = r;
  l !== null && (c = wr({ inputs: { x: r }, attrs: { perm: l }, backend: n }));
  let p = C.getInnerMostAxes(1, u)[0];
  C.assertAxesAreInnerMostDims("cumsum", [p], u);
  let d = n.makeOutput(c.shape, c.dtype), h = c.shape[p], f = n.dataIdMap.get(c.dataId).id, m = n.dataIdMap.get(d.dataId).id;
  cN(f, i ? 1 : 0, o ? 1 : 0, h, m, St[r.dtype]);
  let g = d;
  if (l !== null) {
    let b = C.getUndoAxesPermutation(l);
    g = wr({ inputs: { x: d }, attrs: { perm: b }, backend: n }), n.disposeData(c.dataId), n.disposeData(d.dataId);
  }
  return g;
}
var Lue = { kernelName: Da, backendName: "wasm", setupFunc: zue, kernelFunc: Mue };
var dN;
function Bue(e) {
  dN = e.wasm.cwrap(bo, null, ["number", "number", "number", "array", "number", "array", "array", "number", "number"]);
}
function Vue(e) {
  let { backend: t, inputs: n, attrs: s } = e, { x: r } = n, { blockSize: a, dataFormat: i } = s, o = r.shape[0], u = i === "NHWC" ? r.shape[1] : r.shape[2], l = i === "NHWC" ? r.shape[2] : r.shape[3], c = i === "NHWC" ? r.shape[3] : r.shape[1], p = u * a, d = l * a, h = c / (a * a), f = i === "NHWC" ? [o, p, d, h] : [o, h, p, d], m = t.makeOutput(f, "float32"), b = t.dataIdMap.get(r.dataId).id, y = new Uint8Array(new Int32Array(w.computeStrides(r.shape)).buffer), v = new Uint8Array(new Int32Array(f).buffer), x = new Uint8Array(new Int32Array(w.computeStrides(f)).buffer), k = t.dataIdMap.get(m.dataId).id;
  return dN(b, a, i === "NHWC" ? 1 : 0, y, r.shape.length - 1, v, x, f.length, k), m;
}
var Wue = { kernelName: bo, backendName: "wasm", setupFunc: Bue, kernelFunc: Vue };
var pN;
function Uue(e) {
  pN = e.wasm.cwrap(Fa, null, ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function Gue(e) {
  let { inputs: t, attrs: n, backend: s } = e, { x: r, filter: a } = t, i = s.dataIdMap.get(r.dataId).id, o = s.dataIdMap.get(a.dataId).id, { strides: u, dilations: l, pad: c, dimRoundingMode: p } = n, d = l == null ? [1, 1] : l, h = C.computeConv2DInfo(r.shape, a.shape, u, d, c, p, true), f = h.filterHeight, m = h.filterWidth, g = h.padInfo.top, b = h.padInfo.right, y = h.padInfo.bottom, v = h.padInfo.left, x = h.dilationHeight, k = h.dilationWidth, I = h.strideHeight, $ = h.strideWidth, R = h.inChannels, E = h.outChannels, P = h.padInfo.type === "SAME" ? 1 : 0;
  if (h.dataFormat !== "channelsLast")
    throw new Error(`wasm backend DepthwiseConv2dNative does not support dataFormat:'${h.dataFormat}'. Please use 'channelsLast'.`);
  let A = s.makeOutput(h.outShape, "float32"), O = s.dataIdMap.get(A.dataId).id;
  return pN(i, r.shape[0], r.shape[1], r.shape[2], o, f, m, g, b, y, v, P, x, k, I, $, R, E, O), A;
}
var Hue = { kernelName: Fa, backendName: "wasm", setupFunc: Uue, kernelFunc: Gue };
var que = Xt(Pa);
var jue = false;
var Kue = gn(yo, jue, "bool");
var Xue = Xt(za, "float32");
function rg(e) {
  let { inputs: t, attrs: n, backend: s } = e, { input: r } = t, { dim: a } = n, i = r.shape.length, o = r.shape.slice(), u = a;
  return a < 0 && (w.assert(-(i + 1) <= a, () => `Axis must be in the interval [${-(i + 1)}, ${i}]`), u = i + a + 1), o.splice(u, 0, 1), yn({ inputs: { x: r }, backend: s, attrs: { shape: o } });
}
var Yue = { kernelName: vo, backendName: "wasm", kernelFunc: rg };
function hN(e) {
  let { attrs: { shape: t, value: n, dtype: s }, backend: r } = e, a = r.makeOutput(t, s);
  return r.typedArrayFromHeap(a).fill(n), a;
}
var Que = { kernelName: yl, backendName: "wasm", kernelFunc: hN };
var fN;
function Zue(e) {
  fN = e.wasm.cwrap(wo, null, ["number", "number", "number", "number", "number", "number"]);
}
function Jue(e) {
  let { inputs: t, backend: n } = e, { image: s } = t, r = n.makeOutput(s.shape, s.dtype), a = n.dataIdMap.get(s.dataId).id, i = n.dataIdMap.get(r.dataId).id, [o, u, l, c] = s.shape;
  return fN(a, o, u, l, c, i), r;
}
var ele = { kernelName: wo, backendName: "wasm", kernelFunc: Jue, setupFunc: Zue };
var tle = Xt(Ma);
var nle = false;
var sle = gn(La, nle);
var mN;
function rle(e) {
  mN = e.wasm.cwrap(Ba, null, ["number", "number", "number", "number", "number", "number", "number"]);
}
function ale(e) {
  let { backend: t, inputs: n, attrs: s } = e, { varianceEpsilon: r } = s, { x: a, mean: i, variance: o, offset: u, scale: l } = n, c = t.dataIdMap.get(a.dataId).id, p = t.dataIdMap.get(i.dataId).id, d = t.dataIdMap.get(o.dataId).id, h = u != null ? t.dataIdMap.get(u.dataId).id : 0, f = l != null ? t.dataIdMap.get(l.dataId).id : 0, m = t.makeOutput(a.shape, a.dtype);
  if (w.sizeFromShape(a.shape) === 0)
    return m;
  let g = t.dataIdMap.get(m.dataId).id;
  return mN(c, p, d, h, f, r, g), m;
}
var ile = { kernelName: Ba, backendName: "wasm", setupFunc: rle, kernelFunc: ale };
var gN;
function ole(e) {
  gN = e.wasm.cwrap(ia, null, ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function ule(e) {
  let { inputs: t, attrs: n, backend: s } = e, { x: r, filter: a, bias: i, preluActivationWeights: o } = t, { strides: u, pad: l, dilations: c, dataFormat: p, dimRoundingMode: d, activation: h, leakyreluAlpha: f } = n, m = C.computeConv2DInfo(r.shape, a.shape, u, c, l, d), g = ih[h];
  if (g == null)
    throw new Error(`${h} activation not yet supported for FusedConv2D in the wasm backend.`);
  let b = s.dataIdMap.get(r.dataId).id, y = s.dataIdMap.get(a.dataId).id, v = m.outChannels, x = 0;
  if (i != null) {
    let ne = s.dataIdMap.get(i.dataId);
    if (ne.shape.length !== 1)
      throw new Error(`FusedConv2D only supports rank-1 bias but got rank ${ne.shape.length}.`);
    if (ne.shape[0] !== v)
      throw new Error(`FusedConv2D bias shape (${ne.shape}) does not match the number of output channels (${v})`);
    x = ne.id;
  }
  let k = m.filterHeight, I = m.filterWidth, $ = m.padInfo.top, R = m.padInfo.right, E = m.padInfo.bottom, P = m.padInfo.left, A = m.dilationHeight, O = m.dilationWidth, T = m.strideHeight, z = m.strideWidth, W = m.inChannels, q = m.padInfo.type === "SAME" ? 1 : 0, X = m.batchSize, Y = m.inHeight, Z = m.inWidth;
  if (p !== "NHWC")
    throw new Error(`wasm backend FusedConv2D does not support dataFormat:'${p}'. Please use 'NHWC'.`);
  let te = s.makeOutput(m.outShape, "float32"), J = s.dataIdMap.get(te.dataId).id, se = o == null ? 0 : s.dataIdMap.get(o.dataId).id;
  return gN(b, X, Y, Z, y, k, I, x, $, R, E, P, q, A, O, T, z, W, v, g, se, f || 0, J), te;
}
var lle = { kernelName: ia, backendName: "wasm", setupFunc: ole, kernelFunc: ule };
var bN;
function cle(e) {
  bN = e.wasm.cwrap(oa, null, ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function dle(e) {
  let { inputs: t, attrs: n, backend: s } = e, { x: r, filter: a, bias: i, preluActivationWeights: o } = t, { strides: u, pad: l, dilations: c, dataFormat: p, dimRoundingMode: d, activation: h, leakyreluAlpha: f } = n, m = C.computeConv2DInfo(r.shape, a.shape, u, c, l, d, true), g = ih[h];
  if (g == null)
    throw new Error(`${h} activation not yet supported for FusedDepthwiseConv2D in the wasm backend.`);
  let b = s.dataIdMap.get(r.dataId).id, y = s.dataIdMap.get(a.dataId).id, v = m.outChannels, x = 0;
  if (i != null) {
    let ne = s.dataIdMap.get(i.dataId);
    if (ne.shape.length !== 1)
      throw new Error(`FusedDepthwiseConv2D only supports rank-1 bias but got rank ${ne.shape.length}.`);
    if (ne.shape[0] !== v)
      throw new Error(`FusedDepthwiseConv2D bias shape (${ne.shape}) does not match the number of output channels (${v})`);
    x = ne.id;
  }
  let k = m.filterHeight, I = m.filterWidth, $ = m.padInfo.top, R = m.padInfo.right, E = m.padInfo.bottom, P = m.padInfo.left, A = m.dilationHeight, O = m.dilationWidth, T = m.strideHeight, z = m.strideWidth, W = m.inChannels, q = m.padInfo.type === "SAME" ? 1 : 0, X = m.batchSize, Y = m.inHeight, Z = m.inWidth;
  if (p !== "NHWC")
    throw new Error(`wasm backend FusedDepthwiseConv2D does not support dataFormat:'${p}'. Please use 'NHWC'.`);
  let te = s.makeOutput(m.outShape, "float32"), J = s.dataIdMap.get(te.dataId).id, se = o == null ? 0 : s.dataIdMap.get(o.dataId).id;
  return bN(b, X, Y, Z, y, k, I, x, $, R, E, P, q, A, O, T, z, W, v, g, se, f || 0, J), te;
}
var ple = { kernelName: oa, backendName: "wasm", setupFunc: cle, kernelFunc: dle };
var yN;
function hle(e) {
  yN = e.wasm.cwrap(So, null, ["number", "number", "number", "number", "number", "number", "array", "number"]);
}
function fle(e) {
  let { backend: t, inputs: n } = e, { params: s, indices: r } = n, [a, i, o, u] = Mk.prepareAndValidate(s, r), l = t.makeOutput(a, s.dtype);
  if (i === 0)
    return l;
  let c = r.shape, p = c[c.length - 1], h = t.dataIdMap.get(s.dataId).id, m = t.dataIdMap.get(r.dataId).id, g = new Uint8Array(new Int32Array(u).buffer), b = t.dataIdMap.get(l.dataId).id;
  return yN(h, St[s.dtype], m, i, p, o, g, b), l;
}
var mle = { kernelName: So, backendName: "wasm", setupFunc: hle, kernelFunc: fle };
var vN;
function gle(e) {
  vN = e.wasm.cwrap("Gather", null, ["number", "number", "array", "number", "number", "number", "array", "number"]);
}
function ble(e) {
  let { backend: t, inputs: n, attrs: s } = e, { x: r, indices: a } = n, { axis: i, batchDims: o } = s, u = w.parseAxisParam(i, r.shape)[0], l = t.readSync(a.dataId), c = r.shape[u];
  for (let E = 0; E < l.length; ++E) {
    let P = l[E];
    w.assert(P <= c - 1 && P >= 0, () => `GatherV2: the index value ${P} is not in [0, ${c - 1}]`);
  }
  let p = C.segment_util.collectGatherOpShapeInfo(r, a, u, o), d = yn({ inputs: { x: r }, attrs: { shape: [p.batchSize, p.outerSize, p.dimSize, p.sliceSize] }, backend: t }), h = w.sizeFromShape(a.shape), f = yn({ inputs: { x: a }, attrs: { shape: [p.batchSize, h / p.batchSize] }, backend: t }), m = [p.batchSize, p.outerSize, h / p.batchSize, p.sliceSize], g = t.makeOutput(m, r.dtype);
  if (w.sizeFromShape(r.shape) === 0)
    return g;
  let b = d.shape.length - 1, v = t.dataIdMap.get(d.dataId).id, k = t.dataIdMap.get(f.dataId).id, I = t.dataIdMap.get(g.dataId).id, $ = new Uint8Array(new Int32Array(w.computeStrides(d.shape)).buffer), R = new Uint8Array(new Int32Array(w.computeStrides(m)).buffer);
  return vN(v, St[r.dtype], $, b, k, p.batchSize, R, I), t.disposeData(d.dataId), t.disposeData(f.dataId), g.shape = p.outputShape, g;
}
var yle = { kernelName: ko, backendName: "wasm", setupFunc: gle, kernelFunc: ble };
var vle = false;
var xle = gn(Io, vle, "bool");
var wle = false;
var kle = gn(Va, wle, "bool");
var xN;
function Sle(e) {
  xN = e.wasm.cwrap(Ua, null, ["number", "number", "number", "number"]);
}
function Ile(e) {
  let { inputs: { x: t }, attrs: { alpha: n }, backend: s } = e, r = s.dataIdMap.get(t.dataId).id, a = s.makeOutput(t.shape, "float32");
  if (w.sizeFromShape(t.shape) !== 0) {
    let i = s.dataIdMap.get(a.dataId).id;
    xN(r, St[t.dtype], n, i);
  }
  return a;
}
var Cle = { kernelName: Ua, backendName: "wasm", setupFunc: Sle, kernelFunc: Ile };
var Nle = false;
var Tle = gn(Co, Nle, "bool");
var $le = false;
var _le = gn(No, $le, "bool");
var Ale = Xt(Ga);
var Ele = false;
var Rle = gn(To, Ele, "bool");
var wN;
function Dle(e) {
  wN = e.wasm.cwrap(Ha, null, ["number", "number", "number", "number"]);
}
function Fle(e) {
  let { backend: t, inputs: n, attrs: s } = e, { reductionIndices: r, keepDims: a } = s, { x: i } = n, u = t.dataIdMap.get(i.dataId).id, l = i, { transposed: c, axes: p, originalAxes: d, inputWasTransposed: h } = Or(i, r, t);
  if (h) {
    let v = t.dataIdMap.get(c.dataId).id;
    l = c, u = v;
  }
  let f = l.shape.length;
  C.assertAxesAreInnerMostDims("max", p, f);
  let [m, g] = C.computeOutAndReduceShapes(l.shape, p), b = w.sizeFromShape(g), y = t.makeOutput(m, i.dtype);
  if (w.sizeFromShape(l.shape) !== 0) {
    let v = t.dataIdMap.get(y.dataId).id;
    wN(u, St[i.dtype], b, v);
  }
  if (h && t.disposeData(c.dataId), a) {
    let v = C.expandShapeToKeepDim(y.shape, d);
    y.shape = v;
  }
  return y;
}
var Ole = { kernelName: Ha, backendName: "wasm", setupFunc: Dle, kernelFunc: Fle };
var Ple = false;
var zle = gn(qa, Ple);
var kN;
function Mle(e) {
  kN = e.wasm.cwrap(ja, null, ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function Lle(e) {
  let { inputs: t, attrs: n, backend: s } = e, r = t.x, a = s.dataIdMap.get(r.dataId).id;
  w.assert(r.dtype === "float32", () => `Error in MaxPool: only float32 input is supported. Got ${r.dtype}.`);
  let { filterSize: i, strides: o, pad: u, dimRoundingMode: l } = n, c = C.computePool2DInfo(r.shape, i, o, 1, u, l), p = c.filterHeight, d = c.filterWidth, h = c.padInfo.top, f = c.padInfo.right, m = c.padInfo.bottom, g = c.padInfo.left, b = c.dilationHeight, y = c.dilationWidth, v = c.strideHeight, x = c.strideWidth, k = c.inChannels, I = c.outChannels;
  if (c.dataFormat !== "channelsLast")
    throw new Error(`wasm backend does not support dataFormat:'${c.dataFormat}'. Please use 'channelsLast'.`);
  let $ = s.makeOutput(c.outShape, "float32"), R = s.dataIdMap.get($.dataId).id;
  return kN(a, r.shape[0], r.shape[1], r.shape[2], p, d, h, f, m, g, b, y, v, x, k, I, R), $;
}
var Ble = { kernelName: ja, backendName: "wasm", setupFunc: Mle, kernelFunc: Lle };
var SN;
function Vle(e) {
  SN = e.wasm.cwrap(Ka, null, ["number, number, number"]);
}
function Wle(e) {
  let { backend: t, inputs: n, attrs: s } = e, { axis: r, keepDims: a } = s, { x: i } = n, o = t.dataIdMap.get(i.dataId).id, u = o, l = i, { transposed: c, axes: p, originalAxes: d, inputWasTransposed: h } = Or(i, r, t), f = p;
  if (h) {
    let x = t.dataIdMap.get(c.dataId).id;
    x !== o && (l = c, u = x, f = C.getInnerMostAxes(f.length, l.shape.length));
  }
  C.assertAxesAreInnerMostDims("mean", f, l.shape.length);
  let [m, g] = C.computeOutAndReduceShapes(l.shape, f), b = w.sizeFromShape(g), y = l;
  l.dtype !== "float32" && (y = oc({ backend: t, inputs: { x: l }, attrs: { dtype: "float32" } }), u = t.dataIdMap.get(y.dataId).id);
  let v = t.makeOutput(m, "float32");
  if (w.sizeFromShape(l.shape) !== 0) {
    let x = t.dataIdMap.get(v.dataId).id;
    SN(u, b, x);
  }
  if (h && t.disposeData(c.dataId), a) {
    let x = C.expandShapeToKeepDim(v.shape, d);
    v.shape = x;
  }
  return l.dtype !== "float32" && t.disposeData(y.dataId), v;
}
var Ule = { kernelName: Ka, backendName: "wasm", setupFunc: Vle, kernelFunc: Wle };
var IN;
function Gle(e) {
  IN = e.wasm.cwrap(Xa, null, ["number", "number", "number", "number"]);
}
function Hle(e) {
  let { backend: t, inputs: n, attrs: s } = e, { axis: r, keepDims: a } = s, { x: i } = n, o = t.dataIdMap.get(i.dataId).id, u = o, l = i, { transposed: c, axes: p, originalAxes: d, inputWasTransposed: h } = Or(i, r, t);
  if (h) {
    let v = t.dataIdMap.get(c.dataId).id;
    v !== o && (l = c, u = v);
  }
  let f = l.shape.length;
  C.assertAxesAreInnerMostDims("min", p, f);
  let [m, g] = C.computeOutAndReduceShapes(l.shape, p), b = w.sizeFromShape(g), y = t.makeOutput(m, l.dtype);
  if (w.sizeFromShape(l.shape) !== 0) {
    let v = t.dataIdMap.get(y.dataId).id;
    IN(u, St[i.dtype], b, v);
  }
  if (h && t.disposeData(c.dataId), a) {
    let v = C.expandShapeToKeepDim(y.shape, d);
    y.shape = v;
  }
  return y;
}
var qle = { kernelName: Xa, backendName: "wasm", setupFunc: Gle, kernelFunc: Hle };
var jle = false;
var Kle = gn(Ya, jle);
var CN = ((e) => (e[e.reflect = 0] = "reflect", e[e.symmetric = 1] = "symmetric", e))(CN || {});
var NN;
function Xle(e) {
  NN = e.wasm.cwrap(Qa, null, ["number", "array", "number", "number", "array", "array", "number", "number"]);
}
function Yle(e) {
  let { inputs: { x: t }, backend: n, attrs: { paddings: s, mode: r } } = e, a = s.map((f, m) => f[0] + t.shape[m] + f[1]), i = n.dataIdMap.get(t.dataId).id, o = n.makeOutput(a, t.dtype), u = n.dataIdMap.get(o.dataId).id, l = new Uint8Array(new Int32Array(t.shape).buffer), c = s.map((f) => f[0]), p = s.map((f) => f[1]), d = new Uint8Array(new Int32Array(c).buffer), h = new Uint8Array(new Int32Array(p).buffer);
  return NN(i, l, t.shape.length, St[t.dtype], d, h, CN[r], u), o;
}
var Qle = { kernelName: Qa, backendName: "wasm", kernelFunc: Yle, setupFunc: Xle };
var Zle = true;
var Jle = gn(Za, Zle);
var ece = Xt($o);
function Lv(e, t) {
  let n = new Int32Array(e.wasm.HEAPU8.buffer, t, 4), s = n[0], r = n[1], a = n[2], i = n[3];
  return e.wasm._free(t), { pSelectedIndices: s, selectedSize: r, pSelectedScores: a, pValidOutputs: i };
}
var TN;
function tce(e) {
  TN = e.wasm.cwrap(Ao, "number", ["number", "number", "number", "number", "number"]);
}
function nce(e) {
  let { backend: t, inputs: n, attrs: s } = e, { iouThreshold: r, maxOutputSize: a, scoreThreshold: i } = s, { boxes: o, scores: u } = n, l = t.dataIdMap.get(o.dataId).id, c = t.dataIdMap.get(u.dataId).id, p = TN(l, c, a, r, i), { pSelectedIndices: d, selectedSize: h, pSelectedScores: f, pValidOutputs: m } = Lv(t, p);
  return t.wasm._free(f), t.wasm._free(m), t.makeOutput([h], "int32", d);
}
var sce = { kernelName: Ao, backendName: "wasm", setupFunc: tce, kernelFunc: nce };
var $N;
function rce(e) {
  $N = e.wasm.cwrap(Cl, "number", ["number", "number", "number", "number", "number", "bool"]);
}
function ace(e) {
  let { backend: t, inputs: n, attrs: s } = e, { iouThreshold: r, maxOutputSize: a, scoreThreshold: i, padToMaxOutputSize: o } = s, { boxes: u, scores: l } = n, c = t.dataIdMap.get(u.dataId).id, p = t.dataIdMap.get(l.dataId).id, d = $N(c, p, a, r, i, o), { pSelectedIndices: h, selectedSize: f, pSelectedScores: m, pValidOutputs: g } = Lv(t, d);
  t.wasm._free(m);
  let b = t.makeOutput([f], "int32", h), y = t.makeOutput([], "int32", g);
  return [b, y];
}
var ice = { kernelName: Cl, backendName: "wasm", setupFunc: rce, kernelFunc: ace };
var _N;
function oce(e) {
  _N = e.wasm.cwrap(Eo, "number", ["number", "number", "number", "number", "number", "number"]);
}
function uce(e) {
  let { backend: t, inputs: n, attrs: s } = e, { iouThreshold: r, maxOutputSize: a, scoreThreshold: i, softNmsSigma: o } = s, { boxes: u, scores: l } = n, c = t.dataIdMap.get(u.dataId).id, p = t.dataIdMap.get(l.dataId).id, d = _N(c, p, a, r, i, o), { pSelectedIndices: h, selectedSize: f, pSelectedScores: m, pValidOutputs: g } = Lv(t, d);
  t.wasm._free(g);
  let b = t.makeOutput([f], "int32", h), y = t.makeOutput([f], "float32", m);
  return [b, y];
}
var lce = { kernelName: Eo, backendName: "wasm", setupFunc: oce, kernelFunc: uce };
var cce = false;
var dce = gn(_o, cce, "bool");
var AN;
function pce(e) {
  AN = e.wasm.cwrap(Do, null, ["number", "number", "number", "number", "number"]);
}
function hce(e) {
  let { inputs: t, backend: n, attrs: s } = e, { indices: r } = t, { depth: a, onValue: i, offValue: o } = s, u = n.makeOutput([...r.shape, a], "int32"), l = n.dataIdMap.get(u.dataId).id, p = n.dataIdMap.get(r.dataId).id;
  return AN(p, a, i, o, l), u;
}
var fce = { kernelName: Do, backendName: "wasm", setupFunc: pce, kernelFunc: hce };
function mce(e) {
  let { inputs: { x: t }, backend: n } = e, s = n.makeOutput(t.shape, t.dtype);
  return n.typedArrayFromHeap(s).fill(1), s;
}
var gce = { kernelName: Ro, backendName: "wasm", kernelFunc: mce };
function bce(e) {
  let { inputs: t, backend: n, attrs: s } = e, { axis: r } = s;
  if (t.length === 1)
    return rg({ inputs: { input: t[0] }, backend: n, attrs: { dim: r } });
  let a = t[0].shape, i = t[0].dtype;
  t.forEach((c) => {
    w.assertShapesMatch(a, c.shape, "All tensors passed to stack must have matching shapes"), w.assert(i === c.dtype, () => "All tensors passed to stack must have matching dtypes");
  });
  let o = [], u = t.map((c) => {
    let p = rg({ inputs: { input: c }, backend: n, attrs: { dim: r } });
    return o.push(p), p;
  }), l = rN({ inputs: u, backend: n, attrs: { axis: r } });
  return o.forEach((c) => n.disposeData(c.dataId)), l;
}
var yce = { kernelName: Fo, backendName: "wasm", kernelFunc: bce };
var EN;
function vce(e) {
  EN = e.wasm.cwrap(Ja, null, ["number", "array", "number", "number", "array", "array", "number", "number"]);
}
function xce(e) {
  let { inputs: { x: t }, backend: n, attrs: { paddings: s, constantValue: r } } = e, a = s.map((m, g) => m[0] + t.shape[g] + m[1]);
  if (w.sizeFromShape(t.shape) === 0)
    return hN({ backend: n, attrs: { shape: a, value: r, dtype: t.dtype } });
  let i = n.dataIdMap.get(t.dataId).id, o = n.makeOutput(a, t.dtype), l = n.dataIdMap.get(o.dataId).id, c = new Uint8Array(new Int32Array(t.shape).buffer), p = s.map((m) => m[0]), d = s.map((m) => m[1]), h = new Uint8Array(new Int32Array(p).buffer), f = new Uint8Array(new Int32Array(d).buffer);
  return EN(i, c, t.shape.length, St[t.dtype], h, f, r, l), o;
}
var RN = { kernelName: Ja, backendName: "wasm", kernelFunc: xce, setupFunc: vce };
var wce = false;
var kce = gn(ei, wce);
var DN;
function Sce(e) {
  DN = e.wasm.cwrap(ti, null, ["number", "number", "number"]);
}
function Ice(e) {
  let { inputs: t, backend: n } = e, { x: s, alpha: r } = t, a = n.dataIdMap.get(s.dataId).id, i = n.dataIdMap.get(r.dataId).id, o = a, u = s, l = u;
  u.dtype !== "float32" && (l = oc({ backend: n, inputs: { x: s }, attrs: { dtype: "float32" } }), o = n.dataIdMap.get(l.dataId).id);
  let c = n.makeOutput(s.shape, "float32"), p = n.dataIdMap.get(c.dataId).id;
  return DN(o, i, p), u.dtype !== "float32" && n.disposeData(l.dataId), c;
}
var Cce = { kernelName: ti, backendName: "wasm", setupFunc: Sce, kernelFunc: Ice };
var FN;
function Nce(e) {
  FN = e.wasm.cwrap(ni, null, ["number", "number", "number", "number"]);
}
function Tce(e) {
  let { backend: t, inputs: n, attrs: s } = e, { axis: r, keepDims: a } = s, { x: i } = n, o = t.dataIdMap.get(i.dataId).id, u = o, l = i, { transposed: c, axes: p, originalAxes: d, inputWasTransposed: h } = Or(i, r, t), f = p;
  if (h) {
    let v = t.dataIdMap.get(c.dataId).id;
    v !== o && (l = c, u = v, f = C.getInnerMostAxes(f.length, l.shape.length));
  }
  C.assertAxesAreInnerMostDims("prod", f, l.shape.length);
  let [m, g] = C.computeOutAndReduceShapes(l.shape, f), b = w.sizeFromShape(g), y = t.makeOutput(m, l.dtype);
  if (w.sizeFromShape(l.shape) !== 0) {
    let v = t.dataIdMap.get(y.dataId).id;
    FN(u, b, St[y.dtype], v);
  }
  if (h && t.disposeData(c.dataId), a) {
    let v = C.expandShapeToKeepDim(y.shape, d);
    y.shape = v;
  }
  return y;
}
var $ce = { kernelName: ni, backendName: "wasm", setupFunc: Nce, kernelFunc: Tce };
var _ce = (e) => {
  let { backend: t, attrs: n } = e, { start: s, stop: r, step: a, dtype: i } = n, o = pv(s, r, a, i), u = t.makeOutput([o.length], i);
  return t.typedArrayFromHeap(u).set(o), u;
};
var Ace = { kernelName: Nl, backendName: "wasm", kernelFunc: _ce };
var Ece = true;
var Rce = gn(Oa, Ece);
var Dce = Xt(si);
var Fce = Xt(ai);
var ON;
function Oce(e) {
  ON = e.wasm.cwrap(ri, null, ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function Pce(e) {
  let { backend: t, inputs: n, attrs: s } = e, { images: r } = n, { alignCorners: a, halfPixelCenters: i, size: o } = s, [u, l] = o, [c, p, d, h] = r.shape, f = [c, u, l, h], m = t.dataIdMap.get(r.dataId), g;
  m.dtype !== "float32" && (g = oc({ backend: t, inputs: { x: r }, attrs: { dtype: "float32" } }), m = t.dataIdMap.get(g.dataId));
  let b = m.id, y = t.makeOutput(f, "float32");
  if (w.sizeFromShape(r.shape) === 0)
    return y;
  let v = t.dataIdMap.get(y.dataId).id;
  return ON(b, c, p, d, h, u, l, a ? 1 : 0, i ? 1 : 0, v), g != null && t.disposeData(g.dataId), y;
}
var zce = { kernelName: ri, backendName: "wasm", setupFunc: Oce, kernelFunc: Pce };
var PN;
function Mce(e) {
  PN = e.wasm.cwrap(Po, null, ["number", "array", "number", "array", "number", "number"]);
}
function Lce(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { dims: a } = s, i = w.parseAxisParam(a, r.shape);
  if (r.shape.length === 0)
    return oh({ inputs: { x: r }, backend: n });
  let o = n.makeOutput(r.shape, r.dtype), u = n.dataIdMap.get(r.dataId).id, l = n.dataIdMap.get(o.dataId).id, c = new Uint8Array(new Int32Array(i).buffer), p = new Uint8Array(new Int32Array(r.shape).buffer);
  PN(u, c, i.length, p, r.shape.length, l);
  let d = yn({ inputs: { x: o }, attrs: { shape: r.shape }, backend: n });
  return n.disposeData(o.dataId), d;
}
var Bce = { kernelName: Po, backendName: "wasm", kernelFunc: Lce, setupFunc: Mce };
var zN;
function Vce(e) {
  zN = e.wasm.cwrap(Yo, null, ["number", "number", "number", "number", "number", "number", "number", "number", "array", "number", "number"]);
}
function Wce(e) {
  let { inputs: t, backend: n, attrs: s } = e, { image: r } = t, { radians: a, fillValue: i, center: o } = s, u = n.makeOutput(r.shape, r.dtype), l = n.dataIdMap.get(r.dataId).id, c = n.dataIdMap.get(u.dataId).id, [p, d, h, f] = r.shape, [m, g] = C.getImageCenter(o, d, h), b = i === 0, y = 255, v = typeof i == "number" ? [i, i, i, b ? 0 : y] : [...i, y], x = new Uint8Array(new Int32Array(v).buffer);
  return zN(l, p, d, h, f, a, m, g, x, v.length, c), u;
}
var Uce = { kernelName: Yo, backendName: "wasm", kernelFunc: Wce, setupFunc: Vce };
var Gce = Xt(zo);
var Hce = Xt(ii);
var MN;
function qce(e) {
  MN = e.wasm.cwrap(Mo, null, ["number", "number", "number", "number", "number", "number", "array", "number", "number"]);
}
function jce(e) {
  let { backend: t, inputs: n, attrs: s } = e, { indices: r, updates: a } = n, { shape: i } = s, o = t.makeOutput(i, a.dtype);
  if (w.sizeFromShape(i) === 0)
    return o;
  let { sliceRank: u, numUpdates: l, sliceSize: c, strides: p, outputSize: d } = Bk.calculateShapes(a, r, i), f = t.dataIdMap.get(r.dataId).id, g = t.dataIdMap.get(a.dataId).id, b = new Uint8Array(new Int32Array(p).buffer), y = t.dataIdMap.get(o.dataId).id;
  return MN(f, g, St[a.dtype], u, l, c, b, d, y), o;
}
var Kce = { kernelName: Mo, backendName: "wasm", setupFunc: qce, kernelFunc: jce };
var LN;
function Xce(e) {
  LN = e.wasm.cwrap("SelectV2", null, ["number", "number", "number", "number", "number"]);
}
function Yce(e) {
  let { inputs: t, backend: n } = e, { condition: s, t: r, e: a } = t, i = n.dataIdMap.get(s.dataId).id, o = n.dataIdMap.get(r.dataId).id, u = n.dataIdMap.get(a.dataId).id, l = n.makeOutput(r.shape, r.dtype), c = n.dataIdMap.get(l.dataId).id, p = s.shape.length, d = r.shape.length, h = p === 0 || p > 1 || d === 1 ? 1 : w.sizeFromShape(r.shape.slice(1));
  return LN(i, o, u, h, c), l;
}
var Qce = { kernelName: Lo, backendName: "wasm", kernelFunc: Yce, setupFunc: Xce };
var BN;
function Zce(e) {
  BN = e.wasm.cwrap(ui, null, ["number", "number"]);
}
function Jce(e) {
  let { backend: t, inputs: { x: n } } = e, s = t.dataIdMap.get(n.dataId).id, r = t.makeOutput(n.shape, n.dtype), a = t.dataIdMap.get(r.dataId).id;
  return w.sizeFromShape(r.shape) === 0 || BN(s, a), r;
}
var ede = { kernelName: "Sigmoid", backendName: "wasm", setupFunc: Zce, kernelFunc: Jce };
var tde = Xt(oi);
var VN;
function nde(e) {
  VN = e.wasm.cwrap(di, null, ["number", "number", "number", "number"]);
}
function sde(e) {
  let { backend: t, inputs: { logits: n }, attrs: { dim: s } } = e, r = t.dataIdMap.get(n.dataId).id, a = t.makeOutput(n.shape, n.dtype), i = t.dataIdMap.get(a.dataId).id, o = n.shape[s], u = w.sizeFromShape(n.shape) / o;
  return w.sizeFromShape(a.shape) === 0 || VN(r, i, o, u), a;
}
var rde = { kernelName: di, backendName: "wasm", setupFunc: nde, kernelFunc: sde };
function ade(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, { blockShape: a, paddings: i } = s, o = w.sizeFromShape(a), u = [[0, 0]];
  u.push(...i);
  for (let I = 1 + a.length; I < r.shape.length; ++I)
    u.push([0, 0]);
  let l = RN.kernelFunc({ inputs: { x: r }, backend: n, attrs: { paddings: u, constantValue: 0 } }), c = C.getReshaped(l.shape, a, o, false), p = C.getPermuted(c.length, a.length, false), d = C.getReshapedPermuted(l.shape, a, o, false), m = yn({ inputs: { x: l }, backend: n, attrs: { shape: c } }), y = wr({ inputs: { x: m }, backend: n, attrs: { perm: p } }), k = yn({ inputs: { x: y }, backend: n, attrs: { shape: d } });
  return n.disposeData(l.dataId), n.disposeData(m.dataId), n.disposeData(y.dataId), k;
}
var ide = { kernelName: Wo, backendName: "wasm", kernelFunc: ade };
var WN;
function ode(e) {
  WN = e.wasm.cwrap("SparseFillEmptyRows", "number", ["number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function ude(e) {
  let { backend: t, inputs: n } = e, { indices: s, values: r, denseShape: a, defaultValue: i } = n, o = s.shape[0], u = s.shape[1], l = t.readSync(a.dataId)[0], c = [o + l, u], p = t.dataIdMap.get(s.dataId).id, d = t.dataIdMap.get(r.dataId).id, h = t.dataIdMap.get(i.dataId).id, f = t.makeOutput(c, s.dtype), m = t.dataIdMap.get(f.dataId).id, g = t.makeOutput(c.slice(0, 1), r.dtype), b = t.dataIdMap.get(g.dataId).id, y = t.makeOutput([l], "bool"), v = t.dataIdMap.get(y.dataId).id, x = t.makeOutput([o], s.dtype), k = t.dataIdMap.get(x.dataId).id, I = t.makeOutput([4], "int32"), $ = t.dataIdMap.get(I.dataId).id, R = WN(p, d, St[r.dtype], o, l, u, h, m, b, v, k, $), E = t.readSync(I.dataId), P;
  switch (E[0]) {
    case 1: {
      P = C.getSparseFillEmptyRowsIndicesDenseShapeMismatch(E[1]);
      break;
    }
    case 2: {
      P = C.getSparseFillEmptyRowsNegativeIndexErrorMessage(E[1], E[2]);
      break;
    }
    case 3:
      P = C.getSparseFillEmptyRowsOutOfRangeIndexErrorMessage(E[1], E[2], E[3]);
      break;
    default:
      P = "";
  }
  if (t.disposeData(I.dataId), P)
    throw t.disposeData(f.dataId), t.disposeData(g.dataId), t.disposeData(y.dataId), t.disposeData(x.dataId), new Error(P);
  let A = f, O = g;
  return R !== c[0] && (A = xa({ inputs: { x: f }, attrs: { begin: 0, size: [R, u] }, backend: t }), O = xa({ inputs: { x: g }, attrs: { begin: 0, size: R }, backend: t }), t.disposeData(f.dataId), t.disposeData(g.dataId)), [A, O, y, x];
}
var lde = { kernelName: cp, backendName: "wasm", setupFunc: ode, kernelFunc: ude };
var UN;
function cde(e) {
  UN = e.wasm.cwrap(Rl, null, ["number", "number", "number", "number", "number", "number", "number"]);
}
function dde(e) {
  let { backend: t, inputs: n } = e, { inputIndices: s, inputShape: r, newShape: a } = n;
  if (s.shape.length !== 2)
    throw new Error(`Input indices should be a matrix but received shape
        ${s.shape}`);
  if (r.shape.length !== 1)
    throw new Error(`Input shape should be a vector but received shape
        ${r.shape}`);
  if (a.shape.length !== 1)
    throw new Error(`Target shape should be a vector but received shape ${a.shape}`);
  let i = t.dataIdMap.get(s.dataId).id, o = t.dataIdMap.get(r.dataId).id, u = t.dataIdMap.get(a.dataId).id, l = s.shape[0], c = w.sizeFromShape(a.shape), p = t.makeOutput([l, c], s.dtype), d = t.dataIdMap.get(p.dataId).id, h = t.makeOutput([c], a.dtype), f = t.dataIdMap.get(h.dataId).id, m = t.makeOutput([3], "int32"), g = t.dataIdMap.get(m.dataId).id;
  UN(i, o, u, l, d, f, g);
  let b = t.readSync(m.dataId), y;
  switch (b[0]) {
    case 0: {
      y = C.getSparseReshapeMultipleNegativeOneOutputDimErrorMessage(b[1], b[2]);
      break;
    }
    case 1: {
      y = C.getSparseReshapeNegativeOutputDimErrorMessage(b[1], b[2]);
      break;
    }
    case 2:
      y = C.getSparseReshapeEmptyTensorZeroOutputDimErrorMessage();
      break;
    case 3: {
      let v = Array.from(t.readSync(r.dataId)), x = Array.from(t.readSync(h.dataId));
      y = C.getSparseReshapeInputOutputMultipleErrorMessage(v, x);
      break;
    }
    case 4: {
      let v = Array.from(t.readSync(r.dataId)), x = Array.from(t.readSync(h.dataId));
      y = C.getSparseReshapeInputOutputMismatchErrorMessage(v, x);
      break;
    }
    default:
      y = "";
  }
  if (t.disposeData(m.dataId), y)
    throw t.disposeData(p.dataId), t.disposeData(h.dataId), new Error(y);
  return [p, h];
}
var pde = { kernelName: Rl, backendName: "wasm", setupFunc: cde, kernelFunc: dde };
var GN;
function HN(e) {
  GN = e.wasm.cwrap("SparseSegmentReduction", null, ["number", "number", "number", "number", "number", "number", "number", "number", "number"]);
}
function qN(e, t) {
  let { backend: n, inputs: s } = e, { data: r, indices: a, segmentIds: i } = s, o = a.shape[0], u = n.readSync(i.dataId, o - 1, o)[0], c = o > 0 ? u + 1 : 0;
  if (c < 0)
    throw new Error(C.getSparseSegmentReductionNegativeSegmentIdsErrorMessage());
  let p = r.shape.slice();
  p[0] = c;
  let d = n.dataIdMap.get(r.dataId).id, h = n.dataIdMap.get(a.dataId).id, f = n.dataIdMap.get(i.dataId).id, m = n.makeOutput(p, r.dtype), g = n.dataIdMap.get(m.dataId).id, b = n.makeOutput([4], "int32"), y = n.dataIdMap.get(b.dataId).id;
  GN(d, St[r.dtype], r.shape[0], h, f, g, y, t, 0);
  let v = n.readSync(b.dataId), x;
  switch (v[0]) {
    case 0: {
      x = C.getSparseSegmentReductionNegativeSegmentIdsErrorMessage();
      break;
    }
    case 1: {
      x = C.getSparseSegmentReductionNonIncreasingSegmentIdsErrorMessage();
      break;
    }
    case 2:
      x = C.getSparseSegmentReductionSegmentIdOutOfRangeErrorMessage(v[1], v[2]);
      break;
    case 3:
      x = C.getSparseSegmentReductionIndicesOutOfRangeErrorMessage(v[1], v[2], v[3]);
      break;
    default:
      x = "";
  }
  if (n.disposeData(b.dataId), x)
    throw n.disposeData(m.dataId), new Error(x);
  return m;
}
function hde(e) {
  return qN(e, true);
}
var fde = { kernelName: dp, backendName: "wasm", setupFunc: HN, kernelFunc: hde };
function mde(e) {
  return qN(e, false);
}
var gde = { kernelName: pp, backendName: "wasm", setupFunc: HN, kernelFunc: mde };
function bde(e) {
  let { inputs: t, attrs: n, backend: s } = e, { x: r } = t, { numOrSizeSplits: a, axis: i } = n, o = w.parseAxisParam(i, r.shape)[0], u = C.prepareSplitSize(r, a, o), l = new Array(r.shape.length).fill(0), c = r.shape.slice();
  return u.map((p) => {
    let d = [...c];
    d[o] = p;
    let h = xa({ inputs: { x: r }, attrs: { begin: l, size: d }, backend: s });
    return l[o] += p, h;
  });
}
var yde = { kernelName: Uo, backendName: "wasm", kernelFunc: bde };
var vde = Xt(li);
var xde = Xt(Dl);
var wde = true;
var kde = gn(pi, wde);
var jN;
function Sde(e) {
  jN = e.wasm.cwrap(gi, null, ["number", "number", "number", "number"]);
}
function Ide(e) {
  let { backend: t, inputs: n, attrs: s } = e, { alpha: r } = s, { x: a } = n, i = t.dataIdMap.get(a.dataId).id, o = t.makeOutput(a.shape, a.dtype), u = t.dataIdMap.get(o.dataId).id;
  return jN(i, r, St[a.dtype], u), o;
}
var Cde = { kernelName: gi, backendName: "wasm", setupFunc: Sde, kernelFunc: Ide };
var KN;
function Nde(e) {
  KN = e.wasm.cwrap(Go, null, ["number", "array", "number", "array", "array", "array", "array", "array", "number", "number"]);
}
function Tde(e) {
  let { backend: t, inputs: n, attrs: s } = e, { x: r } = n, { begin: a, end: i, strides: o, beginMask: u, endMask: l, ellipsisMask: c, newAxisMask: p, shrinkAxisMask: d } = s, { finalShapeSparse: h, finalShape: f, isIdentity: m, sliceDim0: g, isSimpleSlice: b, begin: y, end: v, strides: x } = wt.sliceInfo(r.shape, a, i, o, u, l, c, p, d), k;
  if (m)
    k = yn({ inputs: { x: r }, backend: t, attrs: { shape: f } });
  else if (g || b) {
    w.assert(r.shape.length >= 1, () => `Input must have rank at least 1, got: ${r.shape.length}`);
    let I = wt.computeOutShape(y, v, x), $ = xa({ inputs: { x: r }, backend: t, attrs: { begin: y, size: I } });
    k = yn({ inputs: { x: $ }, backend: t, attrs: { shape: f } }), t.disposeData($.dataId);
  } else {
    let I = t.makeOutput(h, "float32"), $ = t.dataIdMap.get(r.dataId).id, R = new Uint8Array(new Int32Array(w.computeStrides(r.shape)).buffer), E = new Uint8Array(new Int32Array(y).buffer), P = new Uint8Array(new Int32Array(v).buffer), A = new Uint8Array(new Int32Array(x).buffer), O = new Uint8Array(new Int32Array(h).buffer), T = new Uint8Array(new Int32Array(w.computeStrides(h)).buffer), z = t.dataIdMap.get(I.dataId).id;
    KN($, R, r.shape.length, E, P, A, O, T, h.length, z), k = yn({ inputs: { x: I }, backend: t, attrs: { shape: f } }), t.disposeData(I.dataId);
  }
  return k;
}
var $de = { kernelName: Go, backendName: "wasm", setupFunc: Nde, kernelFunc: Tde };
var _de = true;
var Ade = gn(hi, _de);
var XN;
function Ede(e) {
  XN = e.wasm.cwrap(ci, null, ["number", "number", "number", "number"]);
}
function Rde(e) {
  let { backend: t, inputs: n, attrs: s } = e, { axis: r, keepDims: a } = s, { x: i } = n, o = t.dataIdMap.get(i.dataId).id, u = o, l = i, { transposed: c, axes: p, originalAxes: d, inputWasTransposed: h } = Or(i, r, t), f = p;
  if (h) {
    let v = t.dataIdMap.get(c.dataId).id;
    v !== o && (l = c, u = v, f = C.getInnerMostAxes(f.length, l.shape.length));
  }
  C.assertAxesAreInnerMostDims("sum", f, l.shape.length);
  let [m, g] = C.computeOutAndReduceShapes(l.shape, f), b = w.sizeFromShape(g), y = t.makeOutput(m, l.dtype);
  if (w.sizeFromShape(l.shape) !== 0) {
    let v = t.dataIdMap.get(y.dataId).id;
    XN(u, b, St[y.dtype], v);
  }
  if (h && t.disposeData(c.dataId), a) {
    let v = C.expandShapeToKeepDim(y.shape, d);
    y.shape = v;
  }
  return y;
}
var Dde = { kernelName: ci, backendName: "wasm", setupFunc: Ede, kernelFunc: Rde };
var Fde = Xt(Ho);
var Ode = Xt(fi);
var YN;
function Pde(e) {
  YN = e.wasm.cwrap(Cr, null, ["number", "array", "number", "array", "number", "number"]);
}
function zde(e) {
  let { inputs: t, backend: n, attrs: s } = e, { x: r } = t, a = n.dataIdMap.get(r.dataId).id, { reps: i } = s, o = new Array(r.shape.length);
  for (let d = 0; d < o.length; d++)
    o[d] = r.shape[d] * i[d];
  let u = new Uint8Array(new Int32Array(r.shape).buffer), l = new Uint8Array(new Int32Array(o).buffer), c = n.makeOutput(o, r.dtype), p = n.dataIdMap.get(c.dataId).id;
  return YN(a, u, r.shape.length, l, o.length, St[c.dtype], p), c;
}
var Mde = { kernelName: Cr, backendName: "wasm", setupFunc: Pde, kernelFunc: zde };
var QN;
function Lde(e) {
  QN = e.wasm.cwrap(qo, null, ["number", "array", "number", "number", "number", "bool", "number", "number"]);
}
var Bde = ({ inputs: e, backend: t, attrs: n }) => {
  let { x: s } = e, { k: r, sorted: a } = n, i = t.dataIdMap.get(s.dataId).id, o = new Uint8Array(new Int32Array(s.shape).buffer), u = s.shape.slice();
  u[u.length - 1] = r;
  let l = t.makeOutput(u, s.dtype), c = t.dataIdMap.get(l.dataId).id, p = t.makeOutput(u, "int32"), d = t.dataIdMap.get(p.dataId).id;
  return QN(i, o, s.shape.length, St[s.dtype], r, a, c, d), [l, p];
};
var Vde = { kernelName: qo, backendName: "wasm", setupFunc: Lde, kernelFunc: Bde };
var ZN;
function Wde(e) {
  ZN = e.wasm.cwrap(jo, null, ["number", "number", "bool", "number", "number", "number", "number", "number", "number", "array", "number", "number", "number", "number", "number"]);
}
function Ude(e) {
  let { backend: t, inputs: n, attrs: s } = e, { image: r, transforms: a } = n, { interpolation: i, fillMode: o, fillValue: u, outputShape: l } = s, [c, p, d, h] = r.shape, [f, m] = l != null ? l : [p, d], g = [c, f, m, h], b = new Uint8Array(new Int32Array(w.computeStrides(r.shape)).buffer), y = t.makeOutput(g, r.dtype), v = t.dataIdMap.get(y.dataId).id, k = t.dataIdMap.get(r.dataId).id, $ = t.dataIdMap.get(a.dataId).id, R = i === "nearest" ? 1 : 2, E;
  switch (o) {
    case "constant":
      E = 1;
      break;
    case "reflect":
      E = 2;
      break;
    case "wrap":
      E = 3;
      break;
    case "nearest":
      E = 4;
      break;
    default:
      E = 1;
      break;
  }
  return ZN(k, $, a.shape[0] > 1, c, f, m, h, d, p, b, r.shape.length - 1, R, E, u, v), y;
}
var Gde = { kernelName: jo, backendName: "wasm", setupFunc: Wde, kernelFunc: Ude };
function Hde(e) {
  let { inputs: t, backend: n, attrs: s } = e, { value: r } = t, { axis: a } = s;
  a < 0 && (a += r.shape.length);
  let i = r.shape[a], o = r.shape.length, u = new Array(o - 1), l = 0;
  for (let h = 0; h < o; h++)
    h !== a && (u[l++] = r.shape[h]);
  let c = new Array(i), p = new Array(o).fill(0), d = r.shape.slice();
  d[a] = 1;
  for (let h = 0; h < c.length; h++)
    p[a] = h, c[h] = xa({ inputs: { x: r }, attrs: { begin: p, size: d }, backend: n });
  return c.map(({ dataId: h, dtype: f }) => ({ dataId: h, dtype: f, shape: u }));
}
var qde = { kernelName: Ko, backendName: "wasm", kernelFunc: Hde };
function jde(e) {
  let { inputs: { x: t }, backend: n } = e, s = n.makeOutput(t.shape, t.dtype);
  return n.typedArrayFromHeap(s).fill(0), s;
}
var Kde = { kernelName: Xo, backendName: "wasm", kernelFunc: jde };
var Xde = [zoe, Moe, Boe, Uoe, Qoe, eue, sue, iue, cue, gue, bue, yue, wue, kue, Cue, $ue, _ue, Aue, Due, Pue, Lue, Wue, Hue, que, Kue, Xue, Yue, Que, ele, tle, sle, ile, lle, ple, mle, yle, xle, kle, Goe, Cle, Tle, _le, Ale, Rle, Ole, zle, Ble, Ule, qle, Kle, Qle, Jle, ece, sce, ice, lce, dce, fce, gce, yce, RN, kce, Cce, $ce, Ace, Rce, Dce, Fce, oue, zce, Bce, Uce, Gce, Hce, Kce, Qce, ede, tde, fue, rde, ide, lde, pde, fde, gde, yde, vde, xde, kde, Cde, $de, Ade, Dde, Fde, Ode, Mde, Vde, Gde, Koe, qde, Kde];
for (let e of Xde)
  Fl(e);
var ag = K();
ag.registerFlag("WASM_HAS_SIMD_SUPPORT", async () => WebAssembly.validate(new Uint8Array([0, 97, 115, 109, 1, 0, 0, 0, 1, 4, 1, 96, 0, 0, 3, 2, 1, 0, 10, 9, 1, 7, 0, 65, 0, 253, 15, 26, 11])));
ag.registerFlag("WASM_HAS_MULTITHREAD_SUPPORT", async () => {
  if (ag.get("IS_NODE"))
    return false;
  try {
    return new MessageChannel().port1.postMessage(new SharedArrayBuffer(1)), WebAssembly.validate(new Uint8Array([0, 97, 115, 109, 1, 0, 0, 0, 1, 4, 1, 96, 0, 0, 3, 2, 1, 0, 5, 4, 1, 3, 1, 1, 10, 11, 1, 9, 0, 65, 0, 254, 16, 2, 0, 26, 11]));
  } catch (e) {
    return false;
  }
});
var Xw = wa(h$());
var Yde = `"use strict";var Module={};var ENVIRONMENT_IS_NODE=typeof process==="object"&&typeof process.versions==="object"&&typeof process.versions.node==="string";if(ENVIRONMENT_IS_NODE){var nodeWorkerThreads=require("worker_threads");var parentPort=nodeWorkerThreads.parentPort;parentPort.on("message",function(data){onmessage({data:data})});var fs=require("fs");Object.assign(global,{self:global,require:require,Module:Module,location:{href:__filename},Worker:nodeWorkerThreads.Worker,importScripts:function(f){(0,eval)(fs.readFileSync(f,"utf8"))},postMessage:function(msg){parentPort.postMessage(msg)},performance:global.performance||{now:function(){return Date.now()}}})}function threadPrintErr(){var text=Array.prototype.slice.call(arguments).join(" ");if(ENVIRONMENT_IS_NODE){fs.writeSync(2,text+"
");return}console.error(text)}function threadAlert(){var text=Array.prototype.slice.call(arguments).join(" ");postMessage({cmd:"alert",text:text,threadId:Module["_pthread_self"]()})}var err=threadPrintErr;self.alert=threadAlert;Module["instantiateWasm"]=((info,receiveInstance)=>{var instance=new WebAssembly.Instance(Module["wasmModule"],info);receiveInstance(instance);Module["wasmModule"]=null;return instance.exports});self.onmessage=(e=>{try{if(e.data.cmd==="load"){Module["wasmModule"]=e.data.wasmModule;Module["wasmMemory"]=e.data.wasmMemory;Module["buffer"]=Module["wasmMemory"].buffer;Module["ENVIRONMENT_IS_PTHREAD"]=true;if(typeof e.data.urlOrBlob==="string"){importScripts(e.data.urlOrBlob)}else{var objectUrl=URL.createObjectURL(e.data.urlOrBlob);importScripts(objectUrl);URL.revokeObjectURL(objectUrl)}WasmBackendModuleThreadedSimd(Module).then(function(instance){Module=instance})}else if(e.data.cmd==="run"){Module["__performance_now_clock_drift"]=performance.now()-e.data.time;Module["__emscripten_thread_init"](e.data.threadInfoStruct,0,0,1);Module["establishStackSpace"]();Module["PThread"].receiveObjectTransfer(e.data);Module["PThread"].threadInit();try{var result=Module["invokeEntryPoint"](e.data.start_routine,e.data.arg);if(Module["keepRuntimeAlive"]()){Module["PThread"].setExitStatus(result)}else{Module["__emscripten_thread_exit"](result)}}catch(ex){if(ex!="unwind"){if(ex instanceof Module["ExitStatus"]){if(Module["keepRuntimeAlive"]()){}else{Module["__emscripten_thread_exit"](ex.status)}}else{throw ex}}}}else if(e.data.cmd==="cancel"){if(Module["_pthread_self"]()){Module["__emscripten_thread_exit"](-1)}}else if(e.data.target==="setimmediate"){}else if(e.data.cmd==="processThreadQueue"){if(Module["_pthread_self"]()){Module["_emscripten_current_thread_process_queued_calls"]()}}else if(e.data.cmd==="processProxyingQueue"){if(Module["_pthread_self"]()){Module["_emscripten_proxy_execute_queue"](e.data.queue)}}else{err("worker.js received unknown command "+e.data.cmd);err(e.data)}}catch(ex){err("worker.js onmessage() captured an uncaught exception: "+ex);if(ex&&ex.stack)err(ex.stack);if(Module["__emscripten_thread_crashed"]){Module["__emscripten_thread_crashed"]()}throw ex}});`;
var Qde = wa(f$());
var Zde = class extends il {
  constructor(e) {
    super(), this.wasm = e, this.dataIdNextNumber = 1, this.wasm.tfjs.initWithThreadsCount(JN), ig = this.wasm.tfjs.getThreadsCount(), this.dataIdMap = new Yd(this, ds());
  }
  write(e, t, n) {
    let s = { id: this.dataIdNextNumber++ };
    return this.move(s, e, t, n, 1), s;
  }
  numDataIds() {
    return this.dataIdMap.numDataIds();
  }
  async time(e) {
    let t = w.now();
    return e(), { kernelMs: w.now() - t };
  }
  move(e, t, n, s, r) {
    let a = this.dataIdNextNumber++;
    if (s === "string") {
      let l = t;
      this.dataIdMap.set(e, { id: a, stringBytes: l, shape: n, dtype: s, memoryOffset: null, refCount: r });
      return;
    }
    let i = w.sizeFromShape(n), o = i * w.bytesPerElement(s), u = this.wasm._malloc(o);
    this.dataIdMap.set(e, { id: a, memoryOffset: u, shape: n, dtype: s, refCount: r }), this.wasm.tfjs.registerTensor(a, i, u), t != null && this.wasm.HEAPU8.set(new Uint8Array(t.buffer, t.byteOffset, o), u);
  }
  async read(e) {
    return this.readSync(e);
  }
  readSync(e, t, n) {
    let { memoryOffset: s, dtype: r, shape: a, stringBytes: i } = this.dataIdMap.get(e);
    if (r === "string")
      return (t == null || t === 0) && (n == null || n >= i.length) ? i : i.slice(t, n);
    t = t || 0, n = n || w.sizeFromShape(a);
    let o = w.bytesPerElement(r), u = this.wasm.HEAPU8.slice(s + t * o, s + n * o);
    return tpe(u.buffer, r);
  }
  disposeData(e, t = false) {
    if (this.dataIdMap.has(e)) {
      let n = this.dataIdMap.get(e);
      if (n.refCount--, !t && n.refCount > 0)
        return false;
      this.wasm._free(n.memoryOffset), this.wasm.tfjs.disposeData(n.id), this.dataIdMap.delete(e);
    }
    return true;
  }
  refCount(e) {
    return this.dataIdMap.has(e) ? this.dataIdMap.get(e).refCount : 0;
  }
  incRef(e) {
    let t = this.dataIdMap.get(e);
    t != null && t.refCount++;
  }
  floatPrecision() {
    return 32;
  }
  getMemoryOffset(e) {
    return this.dataIdMap.get(e).memoryOffset;
  }
  dispose() {
    this.wasm.tfjs.dispose(), "PThread" in this.wasm && this.wasm.PThread.terminateAllThreads(), this.wasm = null;
  }
  memory() {
    return { unreliable: false };
  }
  makeOutput(e, t, n) {
    let s;
    if (n == null)
      s = this.write(null, e, t);
    else {
      let r = this.dataIdNextNumber++;
      s = { id: r }, this.dataIdMap.set(s, { id: r, memoryOffset: n, shape: e, dtype: t, refCount: 1 });
      let a = w.sizeFromShape(e);
      this.wasm.tfjs.registerTensor(r, a, n);
    }
    return { dataId: s, shape: e, dtype: t };
  }
  typedArrayFromHeap({ shape: e, dtype: t, dataId: n }) {
    let s = this.wasm.HEAPU8.buffer, { memoryOffset: r } = this.dataIdMap.get(n), a = w.sizeFromShape(e);
    switch (t) {
      case "float32":
        return new Float32Array(s, r, a);
      case "int32":
        return new Int32Array(s, r, a);
      case "bool":
        return new Uint8Array(s, r, a);
      default:
        throw new Error(`Unknown dtype ${t}`);
    }
  }
};
function Jde(e) {
  return (t, n) => (w.fetch(e, { credentials: "same-origin" }).then((s) => {
    s.ok || t.env.a(`failed to load wasm binary file at '${e}'`), s.arrayBuffer().then((r) => {
      WebAssembly.instantiate(r, t).then((a) => {
        n(a.instance, a.module);
      });
    });
  }), {});
}
function Yw(e, t, n) {
  if (jd != null)
    return jd;
  let s = "tfjs-backend-wasm.wasm";
  return e && t ? s = "tfjs-backend-wasm-threaded-simd.wasm" : e && (s = "tfjs-backend-wasm-simd.wasm"), Uu != null && Uu[s] != null ? Uu[s] : n + s;
}
async function epe() {
  let [e, t] = await Promise.all([K().getAsync("WASM_HAS_SIMD_SUPPORT"), K().getAsync("WASM_HAS_MULTITHREAD_SUPPORT")]);
  return new Promise((n, s) => {
    let r = {};
    r.locateFile = (o, u) => {
      if (o.endsWith(".worker.js")) {
        let l = Yde.replace(/\n/g, "\\n"), c = new Blob([l], { type: "application/javascript" });
        return URL.createObjectURL(c);
      }
      return o.endsWith(".wasm") ? Yw(e, t, Bu != null ? Bu : u) : u + o;
    }, Bv && (r.instantiateWasm = Jde(Yw(e, t, Bu != null ? Bu : "")));
    let a = false;
    r.onAbort = () => {
      if (a || Gu)
        return;
      Gu = true, s({ message: "Make sure the server can serve the `.wasm` file relative to the bundled js file. For more details see https://github.com/tensorflow/tfjs/blob/master/tfjs-backend-wasm/README.md#using-bundlers" });
    };
    let i;
    t && e && jd == null ? (r.mainScriptUrlOrBlob = new Blob(["var WasmBackendModuleThreadedSimd = " + Xw.default.toString()], { type: "text/javascript" }), i = (0, Xw.default)(r)) : i = (0, Qde.default)(r), i.then((o) => {
      a = true, Gu = false;
      let u = null;
      o.tfjs = { init: o.cwrap("init", null, []), initWithThreadsCount: o.cwrap("init_with_threads_count", null, ["number"]), getThreadsCount: o.cwrap("get_threads_count", "number", []), registerTensor: o.cwrap("register_tensor", null, ["number", "number", "number"]), disposeData: o.cwrap("dispose_data", u, ["number"]), dispose: o.cwrap("dispose", u, []) }, n({ wasm: o });
    });
  });
}
function tpe(e, t) {
  switch (t) {
    case "float32":
      return new Float32Array(e);
    case "int32":
      return new Int32Array(e);
    case "bool":
      return new Uint8Array(e);
    default:
      throw new Error(`Unknown dtype ${t}`);
  }
}
var npe = ["tfjs-backend-wasm.wasm", "tfjs-backend-wasm-simd.wasm", "tfjs-backend-wasm-threaded-simd.wasm"];
var jd = null;
var Bu = null;
var Uu = {};
var Gu = false;
var Bv = false;
function bhe(e, t = false) {
  if (eS("setWasmPath has been deprecated in favor of setWasmPaths and will be removed in a future release."), Gu)
    throw new Error("The WASM backend was already initialized. Make sure you call `setWasmPath()` before you call `tf.setBackend()` or `tf.ready()`");
  jd = e, Bv = t;
}
function yhe(e, t = false) {
  if (Gu)
    throw new Error("The WASM backend was already initialized. Make sure you call `setWasmPaths()` before you call `tf.setBackend()` or `tf.ready()`");
  if (typeof e == "string")
    Bu = e;
  else {
    Uu = e;
    let n = npe.filter((s) => Uu[s] == null);
    if (n.length > 0)
      throw new Error(`There were no entries found for the following binaries: ${n.join(",")}. Please either call setWasmPaths with a map providing a path for each binary, or with a string indicating the directory where all the binaries can be found.`);
  }
  Bv = t;
}
var JN = -1;
var ig = -1;
function vhe(e) {
  JN = e;
}
function xhe() {
  if (ig === -1)
    throw new Error("WASM backend not initialized.");
  return ig;
}
var whe = "0.0.0";
var spe = 2;
vp("wasm", async () => {
  let { wasm: e } = await epe();
  return new Zde(e);
}, spe);
var nr = "3.17.0-20220518";
var khe = { tfjs: nr, "tfjs-core": nr, "tfjs-data": nr, "tfjs-layers": nr, "tfjs-converter": nr, "tfjs-backend-cpu": nr, "tfjs-backend-webgl": nr, "tfjs-backend-wasm": nr };

// src/image/imagefxshaders.ts
var vertexIdentity = `
  precision highp float;
  attribute vec2 pos;
  attribute vec2 uv;
  varying vec2 vUv;
  uniform float flipY;
  void main(void) {
    vUv = uv;
    gl_Position = vec4(pos.x, pos.y*flipY, 0.0, 1.);
  }
`;
var colorMatrixWithAlpha = `
  precision highp float;
  varying vec2 vUv;
  uniform sampler2D texture;
  uniform float m[20];
  void main(void) {
    vec4 c = texture2D(texture, vUv);
    gl_FragColor.r = m[0] * c.r + m[1] * c.g + m[2] * c.b + m[3] * c.a + m[4];
    gl_FragColor.g = m[5] * c.r + m[6] * c.g + m[7] * c.b + m[8] * c.a + m[9];
    gl_FragColor.b = m[10] * c.r + m[11] * c.g + m[12] * c.b + m[13] * c.a + m[14];
    gl_FragColor.a = m[15] * c.r + m[16] * c.g + m[17] * c.b + m[18] * c.a + m[19];
  }
`;
var colorMatrixWithoutAlpha = `
  precision highp float;
  varying vec2 vUv;
  uniform sampler2D texture;
  uniform float m[20];
  void main(void) {
    vec4 c = texture2D(texture, vUv);
    gl_FragColor.r = m[0] * c.r + m[1] * c.g + m[2] * c.b + m[4];
    gl_FragColor.g = m[5] * c.r + m[6] * c.g + m[7] * c.b + m[9];
    gl_FragColor.b = m[10] * c.r + m[11] * c.g + m[12] * c.b + m[14];
    gl_FragColor.a = c.a;
  }
`;
var pixelate = `
  precision highp float;
  varying vec2 vUv;
  uniform vec2 size;
  uniform sampler2D texture;
  vec2 pixelate(vec2 coord, vec2 size) {
    return floor( coord / size ) * size;
  }
  void main(void) {
    gl_FragColor = vec4(0.0);
    vec2 coord = pixelate(vUv, size);
    gl_FragColor += texture2D(texture, coord);
  }
`;
var blur = `
  precision highp float;
  varying vec2 vUv;
  uniform sampler2D texture;
  uniform vec2 px;
  void main(void) {
    gl_FragColor = vec4(0.0);
    gl_FragColor += texture2D(texture, vUv + vec2(-7.0*px.x, -7.0*px.y))*0.0044299121055113265;
    gl_FragColor += texture2D(texture, vUv + vec2(-6.0*px.x, -6.0*px.y))*0.00895781211794;
    gl_FragColor += texture2D(texture, vUv + vec2(-5.0*px.x, -5.0*px.y))*0.0215963866053;
    gl_FragColor += texture2D(texture, vUv + vec2(-4.0*px.x, -4.0*px.y))*0.0443683338718;
    gl_FragColor += texture2D(texture, vUv + vec2(-3.0*px.x, -3.0*px.y))*0.0776744219933;
    gl_FragColor += texture2D(texture, vUv + vec2(-2.0*px.x, -2.0*px.y))*0.115876621105;
    gl_FragColor += texture2D(texture, vUv + vec2(-1.0*px.x, -1.0*px.y))*0.147308056121;
    gl_FragColor += texture2D(texture, vUv                             )*0.159576912161;
    gl_FragColor += texture2D(texture, vUv + vec2( 1.0*px.x,  1.0*px.y))*0.147308056121;
    gl_FragColor += texture2D(texture, vUv + vec2( 2.0*px.x,  2.0*px.y))*0.115876621105;
    gl_FragColor += texture2D(texture, vUv + vec2( 3.0*px.x,  3.0*px.y))*0.0776744219933;
    gl_FragColor += texture2D(texture, vUv + vec2( 4.0*px.x,  4.0*px.y))*0.0443683338718;
    gl_FragColor += texture2D(texture, vUv + vec2( 5.0*px.x,  5.0*px.y))*0.0215963866053;
    gl_FragColor += texture2D(texture, vUv + vec2( 6.0*px.x,  6.0*px.y))*0.00895781211794;
    gl_FragColor += texture2D(texture, vUv + vec2( 7.0*px.x,  7.0*px.y))*0.0044299121055113265;
  }
`;
var convolution = `
  precision highp float;
  varying vec2 vUv;
  uniform sampler2D texture;
  uniform vec2 px;
  uniform float m[9];
  void main(void) {
    vec4 c11 = texture2D(texture, vUv - px); // top left
    vec4 c12 = texture2D(texture, vec2(vUv.x, vUv.y - px.y)); // top center
    vec4 c13 = texture2D(texture, vec2(vUv.x + px.x, vUv.y - px.y)); // top right
    vec4 c21 = texture2D(texture, vec2(vUv.x - px.x, vUv.y) ); // mid left
    vec4 c22 = texture2D(texture, vUv); // mid center
    vec4 c23 = texture2D(texture, vec2(vUv.x + px.x, vUv.y) ); // mid right
    vec4 c31 = texture2D(texture, vec2(vUv.x - px.x, vUv.y + px.y) ); // bottom left
    vec4 c32 = texture2D(texture, vec2(vUv.x, vUv.y + px.y) ); // bottom center
    vec4 c33 = texture2D(texture, vUv + px ); // bottom right
    gl_FragColor = 
    c11 * m[0] + c12 * m[1] + c22 * m[2] +
    c21 * m[3] + c22 * m[4] + c23 * m[5] +
    c31 * m[6] + c32 * m[7] + c33 * m[8];
    gl_FragColor.a = c22.a;
  }
`;

// src/image/imagefx.ts
var collect = (source, prefix, collection) => {
  const r = new RegExp("\\b" + prefix + " \\w+ (\\w+)", "ig");
  source.replace(r, (match3, name) => {
    collection[name] = 0;
    return match3;
  });
};
var GLProgram = class {
  constructor(gl2, vertexSource, fragmentSource) {
    __publicField(this, "uniform", {});
    __publicField(this, "attribute", {});
    __publicField(this, "gl");
    __publicField(this, "id");
    __publicField(this, "compile", (source, type) => {
      const shader = this.gl.createShader(type);
      if (!shader) {
        log("filter: could not create shader");
        return null;
      }
      this.gl.shaderSource(shader, source);
      this.gl.compileShader(shader);
      if (!this.gl.getShaderParameter(shader, this.gl.COMPILE_STATUS)) {
        log(`filter: gl compile failed: ${this.gl.getShaderInfoLog(shader)}`);
        return null;
      }
      return shader;
    });
    this.gl = gl2;
    const vertexShader = this.compile(vertexSource, this.gl.VERTEX_SHADER);
    const fragmentShader = this.compile(fragmentSource, this.gl.FRAGMENT_SHADER);
    this.id = this.gl.createProgram();
    if (!vertexShader || !fragmentShader)
      return;
    if (!this.id) {
      log("filter: could not create webgl program");
      return;
    }
    this.gl.attachShader(this.id, vertexShader);
    this.gl.attachShader(this.id, fragmentShader);
    this.gl.linkProgram(this.id);
    if (!this.gl.getProgramParameter(this.id, this.gl.LINK_STATUS)) {
      log(`filter: gl link failed: ${this.gl.getProgramInfoLog(this.id)}`);
      return;
    }
    this.gl.useProgram(this.id);
    collect(vertexSource, "attribute", this.attribute);
    for (const a in this.attribute)
      this.attribute[a] = this.gl.getAttribLocation(this.id, a);
    collect(vertexSource, "uniform", this.uniform);
    collect(fragmentSource, "uniform", this.uniform);
    for (const u in this.uniform)
      this.uniform[u] = this.gl.getUniformLocation(this.id, u);
  }
};
function GLImageFilter() {
  let drawCount = 0;
  let sourceTexture = null;
  let lastInChain = false;
  let currentFramebufferIndex = -1;
  let tempFramebuffers = [null, null];
  let filterChain = [];
  let vertexBuffer = null;
  let currentProgram = null;
  const fxcanvas = canvas(100, 100);
  const shaderProgramCache = {};
  const DRAW = { INTERMEDIATE: 1 };
  const gl2 = fxcanvas.getContext("webgl");
  if (!gl2) {
    log("filter: cannot get webgl context");
    return;
  }
  this.gl = gl2;
  function resize(width, height) {
    if (width === fxcanvas.width && height === fxcanvas.height)
      return;
    fxcanvas.width = width;
    fxcanvas.height = height;
    if (!vertexBuffer) {
      const vertices = new Float32Array([-1, -1, 0, 1, 1, -1, 1, 1, -1, 1, 0, 0, -1, 1, 0, 0, 1, -1, 1, 1, 1, 1, 1, 0]);
      vertexBuffer = gl2.createBuffer();
      gl2.bindBuffer(gl2.ARRAY_BUFFER, vertexBuffer);
      gl2.bufferData(gl2.ARRAY_BUFFER, vertices, gl2.STATIC_DRAW);
      gl2.pixelStorei(gl2.UNPACK_PREMULTIPLY_ALPHA_WEBGL, true);
    }
    gl2.viewport(0, 0, fxcanvas.width, fxcanvas.height);
    tempFramebuffers = [null, null];
  }
  function createFramebufferTexture(width, height) {
    const fbo = gl2.createFramebuffer();
    gl2.bindFramebuffer(gl2.FRAMEBUFFER, fbo);
    const renderbuffer = gl2.createRenderbuffer();
    gl2.bindRenderbuffer(gl2.RENDERBUFFER, renderbuffer);
    const texture = gl2.createTexture();
    gl2.bindTexture(gl2.TEXTURE_2D, texture);
    gl2.texImage2D(gl2.TEXTURE_2D, 0, gl2.RGBA, width, height, 0, gl2.RGBA, gl2.UNSIGNED_BYTE, null);
    gl2.texParameteri(gl2.TEXTURE_2D, gl2.TEXTURE_MAG_FILTER, gl2.LINEAR);
    gl2.texParameteri(gl2.TEXTURE_2D, gl2.TEXTURE_MIN_FILTER, gl2.LINEAR);
    gl2.texParameteri(gl2.TEXTURE_2D, gl2.TEXTURE_WRAP_S, gl2.CLAMP_TO_EDGE);
    gl2.texParameteri(gl2.TEXTURE_2D, gl2.TEXTURE_WRAP_T, gl2.CLAMP_TO_EDGE);
    gl2.framebufferTexture2D(gl2.FRAMEBUFFER, gl2.COLOR_ATTACHMENT0, gl2.TEXTURE_2D, texture, 0);
    gl2.bindTexture(gl2.TEXTURE_2D, null);
    gl2.bindFramebuffer(gl2.FRAMEBUFFER, null);
    return { fbo, texture };
  }
  function getTempFramebuffer(index2) {
    tempFramebuffers[index2] = tempFramebuffers[index2] || createFramebufferTexture(fxcanvas.width, fxcanvas.height);
    return tempFramebuffers[index2];
  }
  function draw(flags = 0) {
    if (!currentProgram)
      return;
    let source = null;
    let target = null;
    let flipY = false;
    if (drawCount === 0)
      source = sourceTexture;
    else
      source = getTempFramebuffer(currentFramebufferIndex).texture || null;
    drawCount++;
    if (lastInChain && !(flags & DRAW.INTERMEDIATE)) {
      target = null;
      flipY = drawCount % 2 === 0;
    } else {
      currentFramebufferIndex = (currentFramebufferIndex + 1) % 2;
      target = getTempFramebuffer(currentFramebufferIndex).fbo || null;
    }
    gl2.bindTexture(gl2.TEXTURE_2D, source);
    gl2.bindFramebuffer(gl2.FRAMEBUFFER, target);
    gl2.uniform1f(currentProgram.uniform["flipY"], flipY ? -1 : 1);
    gl2.drawArrays(gl2.TRIANGLES, 0, 6);
  }
  function compileShader(fragmentSource) {
    if (shaderProgramCache[fragmentSource]) {
      currentProgram = shaderProgramCache[fragmentSource];
      gl2.useProgram((currentProgram ? currentProgram.id : null) || null);
      return currentProgram;
    }
    currentProgram = new GLProgram(gl2, vertexIdentity, fragmentSource);
    if (!currentProgram) {
      log("filter: could not get webgl program");
      return null;
    }
    const floatSize = Float32Array.BYTES_PER_ELEMENT;
    const vertSize = 4 * floatSize;
    gl2.enableVertexAttribArray(currentProgram.attribute["pos"]);
    gl2.vertexAttribPointer(currentProgram.attribute["pos"], 2, gl2.FLOAT, false, vertSize, 0 * floatSize);
    gl2.enableVertexAttribArray(currentProgram.attribute["uv"]);
    gl2.vertexAttribPointer(currentProgram.attribute["uv"], 2, gl2.FLOAT, false, vertSize, 2 * floatSize);
    shaderProgramCache[fragmentSource] = currentProgram;
    return currentProgram;
  }
  const filter = {
    colorMatrix: (matrix) => {
      const m = new Float32Array(matrix);
      m[4] /= 255;
      m[9] /= 255;
      m[14] /= 255;
      m[19] /= 255;
      const shader = m[18] === 1 && m[3] === 0 && m[8] === 0 && m[13] === 0 && m[15] === 0 && m[16] === 0 && m[17] === 0 && m[19] === 0 ? colorMatrixWithoutAlpha : colorMatrixWithAlpha;
      const program = compileShader(shader);
      if (!program)
        return;
      gl2.uniform1fv(program.uniform["m"], m);
      draw();
    },
    brightness: (brightness) => {
      const b = (brightness || 0) + 1;
      filter.colorMatrix([
        b,
        0,
        0,
        0,
        0,
        0,
        b,
        0,
        0,
        0,
        0,
        0,
        b,
        0,
        0,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    saturation: (amount) => {
      const x = (amount || 0) * 2 / 3 + 1;
      const y = (x - 1) * -0.5;
      filter.colorMatrix([
        x,
        y,
        y,
        0,
        0,
        y,
        x,
        y,
        0,
        0,
        y,
        y,
        x,
        0,
        0,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    desaturate: () => {
      filter.saturation(-1);
    },
    contrast: (amount) => {
      const v = (amount || 0) + 1;
      const o = -128 * (v - 1);
      filter.colorMatrix([
        v,
        0,
        0,
        0,
        o,
        0,
        v,
        0,
        0,
        o,
        0,
        0,
        v,
        0,
        o,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    negative: () => {
      filter.contrast(-2);
    },
    hue: (rotation) => {
      rotation = (rotation || 0) / 180 * Math.PI;
      const cos = Math.cos(rotation);
      const sin = Math.sin(rotation);
      const lumR = 0.213;
      const lumG = 0.715;
      const lumB = 0.072;
      filter.colorMatrix([
        lumR + cos * (1 - lumR) + sin * -lumR,
        lumG + cos * -lumG + sin * -lumG,
        lumB + cos * -lumB + sin * (1 - lumB),
        0,
        0,
        lumR + cos * -lumR + sin * 0.143,
        lumG + cos * (1 - lumG) + sin * 0.14,
        lumB + cos * -lumB + sin * -0.283,
        0,
        0,
        lumR + cos * -lumR + sin * -(1 - lumR),
        lumG + cos * -lumG + sin * lumG,
        lumB + cos * (1 - lumB) + sin * lumB,
        0,
        0,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    desaturateLuminance: () => {
      filter.colorMatrix([
        0.2764723,
        0.929708,
        0.0938197,
        0,
        -37.1,
        0.2764723,
        0.929708,
        0.0938197,
        0,
        -37.1,
        0.2764723,
        0.929708,
        0.0938197,
        0,
        -37.1,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    sepia: () => {
      filter.colorMatrix([
        0.393,
        0.7689999,
        0.18899999,
        0,
        0,
        0.349,
        0.6859999,
        0.16799999,
        0,
        0,
        0.272,
        0.5339999,
        0.13099999,
        0,
        0,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    brownie: () => {
      filter.colorMatrix([
        0.5997023498159715,
        0.34553243048391263,
        -0.2708298674538042,
        0,
        47.43192855600873,
        -0.037703249837783157,
        0.8609577587992641,
        0.15059552388459913,
        0,
        -36.96841498319127,
        0.24113635128153335,
        -0.07441037908422492,
        0.44972182064877153,
        0,
        -7.562075277591283,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    vintagePinhole: () => {
      filter.colorMatrix([
        0.6279345635605994,
        0.3202183420819367,
        -0.03965408211312453,
        0,
        9.651285835294123,
        0.02578397704808868,
        0.6441188644374771,
        0.03259127616149294,
        0,
        7.462829176470591,
        0.0466055556782719,
        -0.0851232987247891,
        0.5241648018700465,
        0,
        5.159190588235296,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    kodachrome: () => {
      filter.colorMatrix([
        1.1285582396593525,
        -0.3967382283601348,
        -0.03992559172921793,
        0,
        63.72958762196502,
        -0.16404339962244616,
        1.0835251566291304,
        -0.05498805115633132,
        0,
        24.732407896706203,
        -0.16786010706155763,
        -0.5603416277695248,
        1.6014850761964943,
        0,
        35.62982807460946,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    technicolor: () => {
      filter.colorMatrix([
        1.9125277891456083,
        -0.8545344976951645,
        -0.09155508482755585,
        0,
        11.793603434377337,
        -0.3087833385928097,
        1.7658908555458428,
        -0.10601743074722245,
        0,
        -70.35205161461398,
        -0.231103377548616,
        -0.7501899197440212,
        1.847597816108189,
        0,
        30.950940869491138,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    polaroid: () => {
      filter.colorMatrix([
        1.438,
        -0.062,
        -0.062,
        0,
        0,
        -0.122,
        1.378,
        -0.122,
        0,
        0,
        -0.016,
        -0.016,
        1.483,
        0,
        0,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    shiftToBGR: () => {
      filter.colorMatrix([
        0,
        0,
        1,
        0,
        0,
        0,
        1,
        0,
        0,
        0,
        1,
        0,
        0,
        0,
        0,
        0,
        0,
        0,
        1,
        0
      ]);
    },
    convolution: (matrix) => {
      const m = new Float32Array(matrix);
      const pixelSizeX = 1 / fxcanvas.width;
      const pixelSizeY = 1 / fxcanvas.height;
      const program = compileShader(convolution);
      if (!program)
        return;
      gl2.uniform1fv(program.uniform["m"], m);
      gl2.uniform2f(program.uniform["px"], pixelSizeX, pixelSizeY);
      draw();
    },
    detectEdges: () => {
      filter.convolution.call(this, [
        0,
        1,
        0,
        1,
        -4,
        1,
        0,
        1,
        0
      ]);
    },
    sobelX: () => {
      filter.convolution.call(this, [
        -1,
        0,
        1,
        -2,
        0,
        2,
        -1,
        0,
        1
      ]);
    },
    sobelY: () => {
      filter.convolution.call(this, [
        -1,
        -2,
        -1,
        0,
        0,
        0,
        1,
        2,
        1
      ]);
    },
    sharpen: (amount) => {
      const a = amount || 1;
      filter.convolution.call(this, [
        0,
        -1 * a,
        0,
        -1 * a,
        1 + 4 * a,
        -1 * a,
        0,
        -1 * a,
        0
      ]);
    },
    emboss: (size2) => {
      const s = size2 || 1;
      filter.convolution.call(this, [
        -2 * s,
        -1 * s,
        0,
        -1 * s,
        1,
        1 * s,
        0,
        1 * s,
        2 * s
      ]);
    },
    blur: (size2) => {
      const blurSizeX = size2 / 7 / fxcanvas.width;
      const blurSizeY = size2 / 7 / fxcanvas.height;
      const program = compileShader(blur);
      if (!program)
        return;
      gl2.uniform2f(program.uniform["px"], 0, blurSizeY);
      draw(DRAW.INTERMEDIATE);
      gl2.uniform2f(program.uniform["px"], blurSizeX, 0);
      draw();
    },
    pixelate: (size2) => {
      const blurSizeX = size2 / fxcanvas.width;
      const blurSizeY = size2 / fxcanvas.height;
      const program = compileShader(pixelate);
      if (!program)
        return;
      gl2.uniform2f(program.uniform["size"], blurSizeX, blurSizeY);
      draw();
    }
  };
  this.add = function(name) {
    const args = Array.prototype.slice.call(arguments, 1);
    const func = filter[name];
    filterChain.push({ func, args });
  };
  this.reset = function() {
    filterChain = [];
  };
  this.get = function() {
    return filterChain;
  };
  this.apply = function(image) {
    resize(image.width, image.height);
    drawCount = 0;
    if (!sourceTexture)
      sourceTexture = gl2.createTexture();
    gl2.bindTexture(gl2.TEXTURE_2D, sourceTexture);
    gl2.texParameteri(gl2.TEXTURE_2D, gl2.TEXTURE_WRAP_S, gl2.CLAMP_TO_EDGE);
    gl2.texParameteri(gl2.TEXTURE_2D, gl2.TEXTURE_WRAP_T, gl2.CLAMP_TO_EDGE);
    gl2.texParameteri(gl2.TEXTURE_2D, gl2.TEXTURE_MIN_FILTER, gl2.NEAREST);
    gl2.texParameteri(gl2.TEXTURE_2D, gl2.TEXTURE_MAG_FILTER, gl2.NEAREST);
    gl2.texImage2D(gl2.TEXTURE_2D, 0, gl2.RGBA, gl2.RGBA, gl2.UNSIGNED_BYTE, image);
    for (let i = 0; i < filterChain.length; i++) {
      lastInChain = i === filterChain.length - 1;
      const f = filterChain[i];
      f.func.apply(this, f.args || []);
    }
    return fxcanvas;
  };
  this.draw = function(image) {
    this.add("brightness", 0);
    return this.apply(image);
  };
}

// src/image/enhance.ts
async function histogramEqualization(inputImage) {
  const squeeze = inputImage.shape.length === 4 ? mr(inputImage) : inputImage;
  const channels = Bn(squeeze, 3, 2);
  const min = [wm(channels[0]), wm(channels[1]), wm(channels[2])];
  const max = [As(channels[0]), As(channels[1]), As(channels[2])];
  const absMax = await Promise.all(max.map((channel) => channel.data()));
  const maxValue = 0.99 * Math.max(absMax[0][0], absMax[1][0], absMax[2][0]);
  const sub = [ge(channels[0], min[0]), ge(channels[1], min[1]), ge(channels[2], min[2])];
  const range = [ge(max[0], min[0]), ge(max[1], min[1]), ge(max[2], min[2])];
  const fact = [xe(maxValue, range[0]), xe(maxValue, range[1]), xe(maxValue, range[2])];
  const enh = [V(sub[0], fact[0]), V(sub[1], fact[1]), V(sub[2], fact[2])];
  const rgb2 = es([enh[0], enh[1], enh[2]], 2);
  const reshape = U(rgb2, [1, squeeze.shape[0], squeeze.shape[1], 3]);
  De([...channels, ...min, ...max, ...sub, ...range, ...fact, ...enh, rgb2, squeeze]);
  return reshape;
}

// src/image/image.ts
var maxSize = 3840;
var inCanvas = null;
var outCanvas = null;
var tmpCanvas = null;
var fx2;
var last = {
  inputSum: 0,
  cacheDiff: 1,
  sumMethod: 0,
  inputTensor: void 0
};
function canvas(width, height) {
  let c;
  if (env.browser) {
    if (env.worker) {
      if (typeof OffscreenCanvas === "undefined")
        throw new Error("canvas error: attempted to run in web worker but OffscreenCanvas is not supported");
      c = new OffscreenCanvas(width, height);
    } else {
      if (typeof document === "undefined")
        throw new Error("canvas error: attempted to run in browser but DOM is not defined");
      c = document.createElement("canvas");
      c.width = width;
      c.height = height;
    }
  } else {
    if (typeof env.Canvas !== "undefined")
      c = new env.Canvas(width, height);
    else if (typeof globalThis.Canvas !== "undefined")
      c = new globalThis.Canvas(width, height);
  }
  return c;
}
function copy(input, output) {
  const outputCanvas = output || canvas(input.width, input.height);
  const ctx = outputCanvas.getContext("2d");
  ctx.drawImage(input, 0, 0);
  return outputCanvas;
}
async function process2(input, config3, getTensor = true) {
  if (!input) {
    if (config3.debug)
      log("input error: input is missing");
    return { tensor: null, canvas: null };
  }
  if (!(input instanceof et) && !(typeof Image !== "undefined" && input instanceof Image) && !(typeof env.Canvas !== "undefined" && input instanceof env.Canvas) && !(typeof globalThis.Canvas !== "undefined" && input instanceof globalThis.Canvas) && !(typeof ImageData !== "undefined" && input instanceof ImageData) && !(typeof ImageBitmap !== "undefined" && input instanceof ImageBitmap) && !(typeof HTMLImageElement !== "undefined" && input instanceof HTMLImageElement) && !(typeof HTMLMediaElement !== "undefined" && input instanceof HTMLMediaElement) && !(typeof HTMLVideoElement !== "undefined" && input instanceof HTMLVideoElement) && !(typeof HTMLCanvasElement !== "undefined" && input instanceof HTMLCanvasElement) && !(typeof OffscreenCanvas !== "undefined" && input instanceof OffscreenCanvas)) {
    throw new Error("input error: type is not recognized");
  }
  if (input instanceof et) {
    let tensor = null;
    if (input["isDisposedInternal"])
      throw new Error("input error: attempted to use tensor but it is disposed");
    if (!input["shape"])
      throw new Error("input error: attempted to use tensor without a shape");
    if (input.shape.length === 3) {
      if (input.shape[2] === 3) {
        tensor = Pn(input, 0);
      } else if (input.shape[2] === 4) {
        const rgb2 = mb(input, [0, 0, 0], [-1, -1, 3]);
        tensor = Pn(rgb2, 0);
        De(rgb2);
      }
    } else if (input.shape.length === 4) {
      if (input.shape[3] === 3) {
        tensor = ur(input);
      } else if (input.shape[3] === 4) {
        tensor = Nd(input, [0, 0, 0, 0], [-1, -1, -1, 3]);
      }
    }
    if (tensor == null || tensor.shape.length !== 4 || tensor.shape[0] !== 1 || tensor.shape[3] !== 3)
      throw new Error(`input error: attempted to use tensor with unrecognized shape: ${input["shape"]}`);
    if (tensor.dtype === "int32") {
      const cast = le(tensor, "float32");
      De(tensor);
      tensor = cast;
    }
    return { tensor, canvas: config3.filter.return ? outCanvas : null };
  } else {
    if (typeof input["readyState"] !== "undefined" && input["readyState"] <= 2) {
      if (config3.debug)
        log("input stream is not ready");
      return { tensor: null, canvas: inCanvas };
    }
    const originalWidth = input["naturalWidth"] || input["videoWidth"] || input["width"] || input["shape"] && input["shape"][1] > 0;
    const originalHeight = input["naturalHeight"] || input["videoHeight"] || input["height"] || input["shape"] && input["shape"][2] > 0;
    if (!originalWidth || !originalHeight) {
      if (config3.debug)
        log("cannot determine input dimensions");
      return { tensor: null, canvas: inCanvas };
    }
    let targetWidth = originalWidth;
    let targetHeight = originalHeight;
    if (targetWidth > maxSize) {
      targetWidth = maxSize;
      targetHeight = Math.trunc(targetWidth * originalHeight / originalWidth);
    }
    if (targetHeight > maxSize) {
      targetHeight = maxSize;
      targetWidth = Math.trunc(targetHeight * originalWidth / originalHeight);
    }
    if ((config3.filter.width || 0) > 0)
      targetWidth = config3.filter.width;
    else if ((config3.filter.height || 0) > 0)
      targetWidth = originalWidth * ((config3.filter.height || 0) / originalHeight);
    if ((config3.filter.height || 0) > 0)
      targetHeight = config3.filter.height;
    else if ((config3.filter.width || 0) > 0)
      targetHeight = originalHeight * ((config3.filter.width || 0) / originalWidth);
    if (!targetWidth || !targetHeight)
      throw new Error("input error: cannot determine dimension");
    if (!inCanvas || (inCanvas == null ? void 0 : inCanvas.width) !== targetWidth || (inCanvas == null ? void 0 : inCanvas.height) !== targetHeight)
      inCanvas = canvas(targetWidth, targetHeight);
    const inCtx = inCanvas.getContext("2d");
    if (typeof ImageData !== "undefined" && input instanceof ImageData) {
      inCtx.putImageData(input, 0, 0);
    } else {
      if (config3.filter.flip && typeof inCtx.translate !== "undefined") {
        inCtx.translate(originalWidth, 0);
        inCtx.scale(-1, 1);
        inCtx.drawImage(input, 0, 0, originalWidth, originalHeight, 0, 0, inCanvas == null ? void 0 : inCanvas.width, inCanvas == null ? void 0 : inCanvas.height);
        inCtx.setTransform(1, 0, 0, 1, 0, 0);
      } else {
        inCtx.drawImage(input, 0, 0, originalWidth, originalHeight, 0, 0, inCanvas == null ? void 0 : inCanvas.width, inCanvas == null ? void 0 : inCanvas.height);
      }
    }
    if (!outCanvas || inCanvas.width !== outCanvas.width || (inCanvas == null ? void 0 : inCanvas.height) !== (outCanvas == null ? void 0 : outCanvas.height))
      outCanvas = canvas(inCanvas.width, inCanvas.height);
    if (config3.filter.enabled && env.webgl.supported) {
      if (!fx2)
        fx2 = env.browser ? new GLImageFilter() : null;
      env.filter = !!fx2;
      if (!fx2 || !fx2.add) {
        if (config3.debug)
          log("input process error: cannot initialize filters");
        env.webgl.supported = false;
        config3.filter.enabled = false;
        copy(inCanvas, outCanvas);
      } else {
        fx2.reset();
        if (config3.filter.brightness !== 0)
          fx2.add("brightness", config3.filter.brightness);
        if (config3.filter.contrast !== 0)
          fx2.add("contrast", config3.filter.contrast);
        if (config3.filter.sharpness !== 0)
          fx2.add("sharpen", config3.filter.sharpness);
        if (config3.filter.blur !== 0)
          fx2.add("blur", config3.filter.blur);
        if (config3.filter.saturation !== 0)
          fx2.add("saturation", config3.filter.saturation);
        if (config3.filter.hue !== 0)
          fx2.add("hue", config3.filter.hue);
        if (config3.filter.negative)
          fx2.add("negative");
        if (config3.filter.sepia)
          fx2.add("sepia");
        if (config3.filter.vintage)
          fx2.add("brownie");
        if (config3.filter.sepia)
          fx2.add("sepia");
        if (config3.filter.kodachrome)
          fx2.add("kodachrome");
        if (config3.filter.technicolor)
          fx2.add("technicolor");
        if (config3.filter.polaroid)
          fx2.add("polaroid");
        if (config3.filter.pixelate !== 0)
          fx2.add("pixelate", config3.filter.pixelate);
        if (fx2.get() > 0)
          outCanvas = fx2.apply(inCanvas);
        else
          outCanvas = fx2.draw(inCanvas);
      }
    } else {
      copy(inCanvas, outCanvas);
      if (fx2)
        fx2 = null;
      env.filter = !!fx2;
    }
    if (!getTensor)
      return { tensor: null, canvas: outCanvas };
    if (!outCanvas)
      throw new Error("canvas error: cannot create output");
    let pixels;
    let depth = 3;
    if (typeof ImageData !== "undefined" && input instanceof ImageData || input["data"] && input["width"] && input["height"]) {
      if (env.browser && Pk) {
        pixels = Pk ? Pk.fromPixels(input) : null;
      } else {
        depth = input["data"].length / input["height"] / input["width"];
        const arr = new Uint8Array(input["data"]["buffer"]);
        pixels = ms(arr, [input["height"], input["width"], depth], "int32");
      }
    } else {
      if (!tmpCanvas || outCanvas.width !== tmpCanvas.width || outCanvas.height !== tmpCanvas.height)
        tmpCanvas = canvas(outCanvas.width, outCanvas.height);
      if (Pk && env.browser) {
        if (config3.backend === "webgl" || config3.backend === "humangl" || config3.backend === "webgpu") {
          pixels = Pk.fromPixels(outCanvas);
        } else {
          tmpCanvas = copy(outCanvas);
          pixels = Pk.fromPixels(tmpCanvas);
        }
      } else {
        const tempCanvas = copy(outCanvas);
        const tempCtx = tempCanvas.getContext("2d");
        const tempData = tempCtx.getImageData(0, 0, targetWidth, targetHeight);
        depth = tempData.data.length / targetWidth / targetHeight;
        const arr = new Uint8Array(tempData.data.buffer);
        pixels = ms(arr, [targetWidth, targetHeight, depth]);
      }
    }
    if (depth === 4) {
      const rgb2 = mb(pixels, [0, 0, 0], [-1, -1, 3]);
      De(pixels);
      pixels = rgb2;
    }
    if (!pixels)
      throw new Error("input error: cannot create tensor");
    const casted = le(pixels, "float32");
    const tensor = config3.filter.equalization ? await histogramEqualization(casted) : Pn(casted, 0);
    De([pixels, casted]);
    return { tensor, canvas: config3.filter.return ? outCanvas : null };
  }
}
async function skip(config3, input) {
  let skipFrame = false;
  if (config3.cacheSensitivity === 0 || !input.shape || input.shape.length !== 4 || input.shape[1] > 2048 || input.shape[2] > 2048)
    return skipFrame;
  if (!last.inputTensor) {
    last.inputTensor = ur(input);
  } else if (last.inputTensor.shape[1] !== input.shape[1] || last.inputTensor.shape[2] !== input.shape[2]) {
    De(last.inputTensor);
    last.inputTensor = ur(input);
  } else {
    const t = {};
    t.diff = ge(input, last.inputTensor);
    t.squared = V(t.diff, t.diff);
    t.sum = ve(t.squared);
    const diffSum = await t.sum.data();
    const diffRelative = diffSum[0] / (input.shape[1] || 1) / (input.shape[2] || 1) / 255 / 3;
    De([last.inputTensor, t.diff, t.squared, t.sum]);
    last.inputTensor = ur(input);
    skipFrame = diffRelative <= (config3.cacheSensitivity || 0);
  }
  return skipFrame;
}
async function compare(config3, input1, input2) {
  const t = {};
  if (!input1 || !input2 || input1.shape.length !== 4 || input1.shape.length !== input2.shape.length) {
    if (!config3.debug)
      log("invalid input tensor or tensor shapes do not match:", input1.shape, input2.shape);
    return 0;
  }
  if (input1.shape[0] !== 1 || input2.shape[0] !== 1 || input1.shape[3] !== 3 || input2.shape[3] !== 3) {
    if (!config3.debug)
      log("input tensors must be of shape [1, height, width, 3]:", input1.shape, input2.shape);
    return 0;
  }
  t.input1 = ur(input1);
  t.input2 = input1.shape[1] !== input2.shape[1] || input1.shape[2] !== input2.shape[2] ? jn.resizeBilinear(input2, [input1.shape[1], input1.shape[2]]) : ur(input2);
  t.diff = ge(t.input1, t.input2);
  t.squared = V(t.diff, t.diff);
  t.sum = ve(t.squared);
  const diffSum = await t.sum.data();
  const diffRelative = diffSum[0] / (input1.shape[1] || 1) / (input1.shape[2] || 1) / 255 / 3;
  De([t.input1, t.input2, t.diff, t.squared, t.sum]);
  return diffRelative;
}

// src/util/env.ts
var Env = class {
  constructor() {
    __publicField(this, "browser");
    __publicField(this, "node");
    __publicField(this, "worker");
    __publicField(this, "platform", "");
    __publicField(this, "agent", "");
    __publicField(this, "backends", []);
    __publicField(this, "initial");
    __publicField(this, "filter");
    __publicField(this, "tfjs");
    __publicField(this, "offscreen");
    __publicField(this, "perfadd", false);
    __publicField(this, "wasm", {
      supported: void 0,
      backend: void 0,
      simd: void 0,
      multithread: void 0
    });
    __publicField(this, "webgl", {
      supported: void 0,
      backend: void 0,
      version: void 0,
      renderer: void 0
    });
    __publicField(this, "webgpu", {
      supported: void 0,
      backend: void 0,
      adapter: void 0
    });
    __publicField(this, "cpu", {
      model: void 0,
      flags: []
    });
    __publicField(this, "kernels", []);
    __publicField(this, "Canvas");
    __publicField(this, "Image");
    __publicField(this, "ImageData");
    this.browser = typeof navigator !== "undefined";
    this.node = typeof process !== "undefined" && typeof process.versions !== "undefined" && typeof process.versions.node !== "undefined";
    this.tfjs = { version: khe["tfjs-core"] };
    this.offscreen = typeof OffscreenCanvas !== "undefined";
    this.initial = true;
    this.worker = this.browser && this.offscreen ? typeof WorkerGlobalScope !== "undefined" : void 0;
    if (typeof navigator !== "undefined") {
      const raw = navigator.userAgent.match(/\(([^()]+)\)/g);
      if (raw && raw[0]) {
        const platformMatch = raw[0].match(/\(([^()]+)\)/g);
        this.platform = platformMatch && platformMatch[0] ? platformMatch[0].replace(/\(|\)/g, "") : "";
        this.agent = navigator.userAgent.replace(raw[0], "");
        if (this.platform[1])
          this.agent = this.agent.replace(raw[1], "");
        this.agent = this.agent.replace(/  /g, " ");
      }
    } else if (typeof process !== "undefined") {
      this.platform = `${process.platform} ${process.arch}`;
      this.agent = `NodeJS ${process.version}`;
    }
  }
  async updateBackend() {
    this.backends = Object.keys(ds().registryFactory);
    this.wasm.supported = typeof WebAssembly !== "undefined";
    this.wasm.backend = this.backends.includes("wasm");
    if (this.wasm.supported && this.wasm.backend && vpe() === "wasm") {
      this.wasm.simd = await K().getAsync("WASM_HAS_SIMD_SUPPORT");
      this.wasm.multithread = await K().getAsync("WASM_HAS_MULTITHREAD_SUPPORT");
    }
    const c = canvas(100, 100);
    const ctx = c ? c.getContext("webgl2") : void 0;
    this.webgl.supported = typeof ctx !== "undefined";
    this.webgl.backend = this.backends.includes("webgl");
    if (this.webgl.supported && this.webgl.backend && (vpe() === "webgl" || vpe() === "humangl")) {
      const gl2 = HA().gpgpu !== "undefined" ? await HA().getGPGPUContext().gl : null;
      if (gl2) {
        this.webgl.version = gl2.getParameter(gl2.VERSION);
        this.webgl.renderer = gl2.getParameter(gl2.RENDERER);
      }
    }
    this.webgpu.supported = this.browser && typeof navigator["gpu"] !== "undefined";
    this.webgpu.backend = this.backends.includes("webgpu");
    try {
      if (this.webgpu.supported)
        this.webgpu.adapter = (await navigator["gpu"].requestAdapter()).name;
    } catch (e) {
      this.webgpu.supported = false;
    }
    try {
      this.kernels = am(vpe()).map((kernel) => kernel.kernelName.toLowerCase());
    } catch (e) {
    }
  }
  async updateCPU() {
    const cpu = { model: "", flags: [] };
    if (this.node && this.platform.startsWith("linux")) {
    }
    if (!this["cpu"])
      Object.defineProperty(this, "cpu", { value: cpu });
    else
      this["cpu"] = cpu;
  }
};
var env = new Env();

// src/tfjs/load.ts
var options = {
  cacheModels: false,
  verbose: true,
  debug: false,
  modelBasePath: ""
};
async function httpHandler(url, init2) {
  if (options.debug)
    log("load model fetch:", url, init2);
  return fetch(url, init2);
}
function setModelLoadOptions(config3) {
  options.cacheModels = config3.cacheModels;
  options.verbose = config3.debug;
  options.modelBasePath = config3.modelBasePath;
}
async function loadModel(modelPath) {
  const modelUrl = join(options.modelBasePath, modelPath || "");
  const modelPathSegments = modelUrl.split("/");
  const cachedModelName = "indexeddb://" + modelPathSegments[modelPathSegments.length - 1].replace(".json", "");
  const cachedModels = await An.listModels();
  const modelCached = options.cacheModels && Object.keys(cachedModels).includes(cachedModelName);
  const tfLoadOptions = typeof fetch === "undefined" ? {} : { fetchFunc: (url, init2) => httpHandler(url, init2) };
  const model18 = new K4(modelCached ? cachedModelName : modelUrl, tfLoadOptions);
  let loaded = false;
  try {
    model18.findIOHandler();
    if (options.debug)
      log("model load handler:", model18.handler);
    const artifacts = await model18.handler.load();
    model18.loadSync(artifacts);
    if (options.verbose)
      log("load model:", model18["modelUrl"]);
    loaded = true;
  } catch (err) {
    log("error loading model:", modelUrl, err);
  }
  if (loaded && options.cacheModels && !modelCached) {
    try {
      const saveResult = await model18.save(cachedModelName);
      log("model saved:", cachedModelName, saveResult);
    } catch (err) {
      log("error saving model:", modelUrl, err);
    }
  }
  return model18;
}

// package.json
var version = "2.7.2";

// src/models.ts
var models_exports = {};
__export(models_exports, {
  Models: () => Models,
  load: () => load19,
  reset: () => reset,
  validate: () => validate2
});

// src/gear/gear.ts
var model;
var last2 = [];
var raceNames = ["white", "black", "asian", "indian", "other"];
var ageWeights = [15, 23, 28, 35.5, 45.5, 55.5, 65];
var lastCount = 0;
var lastTime = 0;
var skipped = Number.MAX_SAFE_INTEGER;
async function load(config3) {
  if (env.initial)
    model = null;
  if (!model)
    model = await loadModel(config3.face["gear"]);
  else if (config3.debug)
    log("cached model:", model["modelUrl"]);
  return model;
}
async function predict(image, config3, idx, count2) {
  var _a2, _b2;
  if (!model)
    return { age: 0, gender: "unknown", genderScore: 0, race: [] };
  const skipFrame = skipped < (((_a2 = config3.face["gear"]) == null ? void 0 : _a2.skipFrames) || 0);
  const skipTime = (((_b2 = config3.face["gear"]) == null ? void 0 : _b2.skipTime) || 0) > now() - lastTime;
  if (config3.skipAllowed && skipTime && skipFrame && lastCount === count2 && last2[idx]) {
    skipped++;
    return last2[idx];
  }
  skipped = 0;
  return new Promise(async (resolve) => {
    var _a3, _b3;
    if (!(model == null ? void 0 : model.inputs[0].shape))
      return;
    const t = {};
    const box = [[0, 0.1, 0.9, 0.9]];
    t.resize = jn.cropAndResize(image, box, [0], [model.inputs[0].shape[2], model.inputs[0].shape[1]]);
    const obj = { age: 0, gender: "unknown", genderScore: 0, race: [] };
    if ((_a3 = config3.face["gear"]) == null ? void 0 : _a3.enabled)
      [t.age, t.gender, t.race] = model.execute(t.resize, ["age_output", "gender_output", "race_output"]);
    const gender = await t.gender.data();
    obj.gender = gender[0] > gender[1] ? "male" : "female";
    obj.genderScore = Math.round(100 * (gender[0] > gender[1] ? gender[0] : gender[1])) / 100;
    const race = await t.race.data();
    for (let i = 0; i < race.length; i++) {
      if (race[i] > (((_b3 = config3.face["gear"]) == null ? void 0 : _b3.minConfidence) || 0.2))
        obj.race.push({ score: Math.round(100 * race[i]) / 100, race: raceNames[i] });
    }
    obj.race.sort((a, b) => b.score - a.score);
    const ageDistribution = Array.from(await t.age.data());
    const ageSorted = ageDistribution.map((a, i) => [ageWeights[i], a]).sort((a, b) => b[1] - a[1]);
    let age = ageSorted[0][0];
    for (let i = 1; i < ageSorted.length; i++)
      age += ageSorted[i][1] * (ageSorted[i][0] - age);
    obj.age = Math.round(10 * age) / 10;
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    last2[idx] = obj;
    lastCount = count2;
    lastTime = now();
    resolve(obj);
  });
}

// src/tfjs/constants.ts
var constants = {
  tf255: 255,
  tf1: 1,
  tf2: 2,
  tf05: 0.5,
  tf127: 127.5,
  rgb: [0.2989, 0.587, 0.114]
};
function init() {
  constants.tf255 = we(255, "float32");
  constants.tf1 = we(1, "float32");
  constants.tf2 = we(2, "float32");
  constants.tf05 = we(0.5, "float32");
  constants.tf127 = we(127.5, "float32");
  constants.rgb = Zt([0.2989, 0.587, 0.114], "float32");
}

// src/gear/ssrnet-age.ts
var model2;
var last3 = [];
var lastCount2 = 0;
var lastTime2 = 0;
var skipped2 = Number.MAX_SAFE_INTEGER;
async function load2(config3) {
  if (env.initial)
    model2 = null;
  if (!model2)
    model2 = await loadModel(config3.face["ssrnet"].modelPathAge);
  else if (config3.debug)
    log("cached model:", model2["modelUrl"]);
  return model2;
}
async function predict2(image, config3, idx, count2) {
  var _a2, _b2, _c, _d2;
  if (!model2)
    return { age: 0 };
  const skipFrame = skipped2 < (((_a2 = config3.face["ssrnet"]) == null ? void 0 : _a2.skipFrames) || 0);
  const skipTime = (((_b2 = config3.face["ssrnet"]) == null ? void 0 : _b2.skipTime) || 0) > now() - lastTime2;
  if (config3.skipAllowed && skipFrame && skipTime && lastCount2 === count2 && ((_c = last3[idx]) == null ? void 0 : _c.age) && ((_d2 = last3[idx]) == null ? void 0 : _d2.age) > 0) {
    skipped2++;
    return last3[idx];
  }
  skipped2 = 0;
  return new Promise(async (resolve) => {
    if (!(model2 == null ? void 0 : model2.inputs) || !model2.inputs[0] || !model2.inputs[0].shape)
      return;
    const t = {};
    t.resize = jn.resizeBilinear(image, [model2.inputs[0].shape[2], model2.inputs[0].shape[1]], false);
    t.enhance = V(t.resize, constants.tf255);
    const obj = { age: 0 };
    if (config3.face["ssrnet"].enabled)
      t.age = model2.execute(t.enhance);
    if (t.age) {
      const data = await t.age.data();
      obj.age = Math.trunc(10 * data[0]) / 10;
    }
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    last3[idx] = obj;
    lastCount2 = count2;
    lastTime2 = now();
    resolve(obj);
  });
}

// src/gear/ssrnet-gender.ts
var model3;
var last4 = [];
var lastCount3 = 0;
var lastTime3 = 0;
var skipped3 = Number.MAX_SAFE_INTEGER;
var rgb = [0.2989, 0.587, 0.114];
async function load3(config3) {
  if (env.initial)
    model3 = null;
  if (!model3)
    model3 = await loadModel(config3.face["ssrnet"].modelPathGender);
  else if (config3.debug)
    log("cached model:", model3["modelUrl"]);
  return model3;
}
async function predict3(image, config3, idx, count2) {
  var _a2, _b2, _c, _d2;
  if (!model3)
    return { gender: "unknown", genderScore: 0 };
  const skipFrame = skipped3 < (((_a2 = config3.face["ssrnet"]) == null ? void 0 : _a2.skipFrames) || 0);
  const skipTime = (((_b2 = config3.face["ssrnet"]) == null ? void 0 : _b2.skipTime) || 0) > now() - lastTime3;
  if (config3.skipAllowed && skipFrame && skipTime && lastCount3 === count2 && ((_c = last4[idx]) == null ? void 0 : _c.gender) && ((_d2 = last4[idx]) == null ? void 0 : _d2.genderScore) > 0) {
    skipped3++;
    return last4[idx];
  }
  skipped3 = 0;
  return new Promise(async (resolve) => {
    if (!(model3 == null ? void 0 : model3.inputs[0].shape))
      return;
    const t = {};
    t.resize = jn.resizeBilinear(image, [model3.inputs[0].shape[2], model3.inputs[0].shape[1]], false);
    t.enhance = j(() => {
      const [red, green, blue] = Bn(t.resize, 3, 3);
      const redNorm = V(red, rgb[0]);
      const greenNorm = V(green, rgb[1]);
      const blueNorm = V(blue, rgb[2]);
      const grayscale = nE([redNorm, greenNorm, blueNorm]);
      const normalize = V(ge(grayscale, constants.tf05), 2);
      return normalize;
    });
    const obj = { gender: "unknown", genderScore: 0 };
    if (config3.face["ssrnet"].enabled)
      t.gender = model3.execute(t.enhance);
    const data = await t.gender.data();
    obj.gender = data[0] > data[1] ? "female" : "male";
    obj.genderScore = data[0] > data[1] ? Math.trunc(100 * data[0]) / 100 : Math.trunc(100 * data[1]) / 100;
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    last4[idx] = obj;
    lastCount3 = count2;
    lastTime3 = now();
    resolve(obj);
  });
}

// src/face/antispoof.ts
var model4;
var cached = [];
var skipped4 = Number.MAX_SAFE_INTEGER;
var lastCount4 = 0;
var lastTime4 = 0;
async function load4(config3) {
  var _a2;
  if (env.initial)
    model4 = null;
  if (!model4)
    model4 = await loadModel((_a2 = config3.face.antispoof) == null ? void 0 : _a2.modelPath);
  else if (config3.debug)
    log("cached model:", model4["modelUrl"]);
  return model4;
}
async function predict4(image, config3, idx, count2) {
  var _a2, _b2;
  if (!model4)
    return 0;
  const skipTime = (((_a2 = config3.face.antispoof) == null ? void 0 : _a2.skipTime) || 0) > now() - lastTime4;
  const skipFrame = skipped4 < (((_b2 = config3.face.antispoof) == null ? void 0 : _b2.skipFrames) || 0);
  if (config3.skipAllowed && skipTime && skipFrame && lastCount4 === count2 && cached[idx]) {
    skipped4++;
    return cached[idx];
  }
  skipped4 = 0;
  return new Promise(async (resolve) => {
    const resize = jn.resizeBilinear(image, [(model4 == null ? void 0 : model4.inputs[0].shape) ? model4.inputs[0].shape[2] : 0, (model4 == null ? void 0 : model4.inputs[0].shape) ? model4.inputs[0].shape[1] : 0], false);
    const res = model4 == null ? void 0 : model4.execute(resize);
    const num = (await res.data())[0];
    cached[idx] = Math.round(100 * num) / 100;
    lastCount4 = count2;
    lastTime4 = now();
    De([resize, res]);
    resolve(cached[idx]);
  });
}

// src/face/facemeshcoords.ts
var meshAnnotations = {
  silhouette: [
    10,
    338,
    297,
    332,
    284,
    251,
    389,
    356,
    454,
    323,
    361,
    288,
    397,
    365,
    379,
    378,
    400,
    377,
    152,
    148,
    176,
    149,
    150,
    136,
    172,
    58,
    132,
    93,
    234,
    127,
    162,
    21,
    54,
    103,
    67,
    109
  ],
  lipsUpperOuter: [61, 185, 40, 39, 37, 0, 267, 269, 270, 409, 291],
  lipsLowerOuter: [146, 91, 181, 84, 17, 314, 405, 321, 375, 291],
  lipsUpperInner: [78, 191, 80, 81, 82, 13, 312, 311, 310, 415, 308],
  lipsLowerInner: [78, 95, 88, 178, 87, 14, 317, 402, 318, 324, 308],
  rightEyeUpper0: [246, 161, 160, 159, 158, 157, 173],
  rightEyeLower0: [33, 7, 163, 144, 145, 153, 154, 155, 133],
  rightEyeUpper1: [247, 30, 29, 27, 28, 56, 190],
  rightEyeLower1: [130, 25, 110, 24, 23, 22, 26, 112, 243],
  rightEyeUpper2: [113, 225, 224, 223, 222, 221, 189],
  rightEyeLower2: [226, 31, 228, 229, 230, 231, 232, 233, 244],
  rightEyeLower3: [143, 111, 117, 118, 119, 120, 121, 128, 245],
  rightEyebrowUpper: [156, 70, 63, 105, 66, 107, 55, 193],
  rightEyebrowLower: [35, 124, 46, 53, 52, 65],
  rightEyeIris: [473, 474, 475, 476, 477],
  leftEyeUpper0: [466, 388, 387, 386, 385, 384, 398],
  leftEyeLower0: [263, 249, 390, 373, 374, 380, 381, 382, 362],
  leftEyeUpper1: [467, 260, 259, 257, 258, 286, 414],
  leftEyeLower1: [359, 255, 339, 254, 253, 252, 256, 341, 463],
  leftEyeUpper2: [342, 445, 444, 443, 442, 441, 413],
  leftEyeLower2: [446, 261, 448, 449, 450, 451, 452, 453, 464],
  leftEyeLower3: [372, 340, 346, 347, 348, 349, 350, 357, 465],
  leftEyebrowUpper: [383, 300, 293, 334, 296, 336, 285, 417],
  leftEyebrowLower: [265, 353, 276, 283, 282, 295],
  leftEyeIris: [468, 469, 470, 471, 472],
  midwayBetweenEyes: [168],
  noseTip: [1],
  noseBottom: [2],
  noseRightCorner: [98],
  noseLeftCorner: [327],
  rightCheek: [205],
  leftCheek: [425]
};
var meshLandmarks = {
  count: 468,
  mouth: 13,
  symmetryLine: [13, meshAnnotations["midwayBetweenEyes"][0]]
};
var blazeFaceLandmarks = {
  leftEye: 0,
  rightEye: 1,
  nose: 2,
  mouth: 3,
  leftEar: 4,
  rightEar: 5,
  symmetryLine: [3, 2]
};
var irisIndices = [
  { key: "EyeUpper0", indices: [9, 10, 11, 12, 13, 14, 15] },
  { key: "EyeUpper1", indices: [25, 26, 27, 28, 29, 30, 31] },
  { key: "EyeUpper2", indices: [41, 42, 43, 44, 45, 46, 47] },
  { key: "EyeLower0", indices: [0, 1, 2, 3, 4, 5, 6, 7, 8] },
  { key: "EyeLower1", indices: [16, 17, 18, 19, 20, 21, 22, 23, 24] },
  { key: "EyeLower2", indices: [32, 33, 34, 35, 36, 37, 38, 39, 40] },
  { key: "EyeLower3", indices: [54, 55, 56, 57, 58, 59, 60, 61, 62] },
  { key: "EyebrowUpper", indices: [63, 64, 65, 66, 67, 68, 69, 70] },
  { key: "EyebrowLower", indices: [48, 49, 50, 51, 52, 53] }
];
var UV468 = [
  [0.499976992607117, 0.652534008026123],
  [0.500025987625122, 0.547487020492554],
  [0.499974012374878, 0.602371990680695],
  [0.482113003730774, 0.471979022026062],
  [0.500150978565216, 0.527155995368958],
  [0.499909996986389, 0.498252987861633],
  [0.499523013830185, 0.40106201171875],
  [0.289712011814117, 0.380764007568359],
  [0.499954998493195, 0.312398016452789],
  [0.499987006187439, 0.269918978214264],
  [0.500023007392883, 0.107050001621246],
  [0.500023007392883, 0.666234016418457],
  [0.5000159740448, 0.679224014282227],
  [0.500023007392883, 0.692348003387451],
  [0.499976992607117, 0.695277988910675],
  [0.499976992607117, 0.70593398809433],
  [0.499976992607117, 0.719385027885437],
  [0.499976992607117, 0.737019002437592],
  [0.499967992305756, 0.781370997428894],
  [0.499816000461578, 0.562981009483337],
  [0.473773002624512, 0.573909997940063],
  [0.104906998574734, 0.254140973091125],
  [0.365929991006851, 0.409575998783112],
  [0.338757991790771, 0.41302502155304],
  [0.311120003461838, 0.409460008144379],
  [0.274657994508743, 0.389131009578705],
  [0.393361985683441, 0.403706014156342],
  [0.345234006643295, 0.344011008739471],
  [0.370094001293182, 0.346076011657715],
  [0.319321990013123, 0.347265005111694],
  [0.297903001308441, 0.353591024875641],
  [0.24779200553894, 0.410809993743896],
  [0.396889001131058, 0.842755019664764],
  [0.280097991228104, 0.375599980354309],
  [0.106310002505779, 0.399955987930298],
  [0.2099249958992, 0.391353011131287],
  [0.355807989835739, 0.534406006336212],
  [0.471751004457474, 0.65040397644043],
  [0.474155008792877, 0.680191993713379],
  [0.439785003662109, 0.657229006290436],
  [0.414617002010345, 0.66654098033905],
  [0.450374007225037, 0.680860996246338],
  [0.428770989179611, 0.682690978050232],
  [0.374971002340317, 0.727805018424988],
  [0.486716985702515, 0.547628998756409],
  [0.485300987958908, 0.527395009994507],
  [0.257764995098114, 0.314490020275116],
  [0.401223003864288, 0.455172002315521],
  [0.429818987846375, 0.548614978790283],
  [0.421351999044418, 0.533740997314453],
  [0.276895999908447, 0.532056987285614],
  [0.483370006084442, 0.499586999416351],
  [0.33721199631691, 0.282882988452911],
  [0.296391993761063, 0.293242990970612],
  [0.169294998049736, 0.193813979625702],
  [0.447580009698868, 0.302609980106354],
  [0.392390012741089, 0.353887975215912],
  [0.354490011930466, 0.696784019470215],
  [0.067304998636246, 0.730105042457581],
  [0.442739009857178, 0.572826027870178],
  [0.457098007202148, 0.584792017936707],
  [0.381974011659622, 0.694710969924927],
  [0.392388999462128, 0.694203019142151],
  [0.277076005935669, 0.271932005882263],
  [0.422551989555359, 0.563233017921448],
  [0.385919004678726, 0.281364023685455],
  [0.383103013038635, 0.255840003490448],
  [0.331431001424789, 0.119714021682739],
  [0.229923993349075, 0.232002973556519],
  [0.364500999450684, 0.189113974571228],
  [0.229622006416321, 0.299540996551514],
  [0.173287004232407, 0.278747975826263],
  [0.472878992557526, 0.666198015213013],
  [0.446828007698059, 0.668527007102966],
  [0.422762006521225, 0.673889994621277],
  [0.445307999849319, 0.580065965652466],
  [0.388103008270264, 0.693961024284363],
  [0.403039008378983, 0.706539988517761],
  [0.403629004955292, 0.693953037261963],
  [0.460041999816895, 0.557139039039612],
  [0.431158006191254, 0.692366003990173],
  [0.452181994915009, 0.692366003990173],
  [0.475387006998062, 0.692366003990173],
  [0.465828001499176, 0.779190003871918],
  [0.472328990697861, 0.736225962638855],
  [0.473087012767792, 0.717857003211975],
  [0.473122000694275, 0.704625964164734],
  [0.473033010959625, 0.695277988910675],
  [0.427942007780075, 0.695277988910675],
  [0.426479011774063, 0.703539967536926],
  [0.423162013292313, 0.711845993995667],
  [0.4183090031147, 0.720062971115112],
  [0.390094995498657, 0.639572978019714],
  [0.013953999616206, 0.560034036636353],
  [0.499913990497589, 0.58014702796936],
  [0.413199990987778, 0.69539999961853],
  [0.409626007080078, 0.701822996139526],
  [0.468080013990402, 0.601534962654114],
  [0.422728985548019, 0.585985004901886],
  [0.463079988956451, 0.593783974647522],
  [0.37211999297142, 0.47341400384903],
  [0.334562003612518, 0.496073007583618],
  [0.411671012639999, 0.546965003013611],
  [0.242175996303558, 0.14767599105835],
  [0.290776997804642, 0.201445996761322],
  [0.327338010072708, 0.256527006626129],
  [0.399509996175766, 0.748921036720276],
  [0.441727995872498, 0.261676013469696],
  [0.429764986038208, 0.187834024429321],
  [0.412198007106781, 0.108901023864746],
  [0.288955003023148, 0.398952007293701],
  [0.218936994671822, 0.435410976409912],
  [0.41278201341629, 0.398970007896423],
  [0.257135003805161, 0.355440020561218],
  [0.427684992551804, 0.437960982322693],
  [0.448339998722076, 0.536936044692993],
  [0.178560003638268, 0.45755398273468],
  [0.247308000922203, 0.457193970680237],
  [0.286267012357712, 0.467674970626831],
  [0.332827985286713, 0.460712015628815],
  [0.368755996227264, 0.447206974029541],
  [0.398963987827301, 0.432654976844788],
  [0.476410001516342, 0.405806005001068],
  [0.189241006970406, 0.523923993110657],
  [0.228962004184723, 0.348950982093811],
  [0.490725994110107, 0.562400996685028],
  [0.404670000076294, 0.485132992267609],
  [0.019469000399113, 0.401564002037048],
  [0.426243007183075, 0.420431017875671],
  [0.396993011236191, 0.548797011375427],
  [0.266469985246658, 0.376977026462555],
  [0.439121007919312, 0.51895797252655],
  [0.032313998788595, 0.644356966018677],
  [0.419054001569748, 0.387154996395111],
  [0.462783008813858, 0.505746960639954],
  [0.238978996872902, 0.779744982719421],
  [0.198220998048782, 0.831938028335571],
  [0.107550002634525, 0.540755033493042],
  [0.183610007166862, 0.740257024765015],
  [0.134409993886948, 0.333683013916016],
  [0.385764002799988, 0.883153975009918],
  [0.490967005491257, 0.579378008842468],
  [0.382384985685349, 0.508572995662689],
  [0.174399003386497, 0.397670984268188],
  [0.318785011768341, 0.39623498916626],
  [0.343364000320435, 0.400596976280212],
  [0.396100014448166, 0.710216999053955],
  [0.187885001301765, 0.588537991046906],
  [0.430987000465393, 0.944064974784851],
  [0.318993002176285, 0.898285031318665],
  [0.266247987747192, 0.869701027870178],
  [0.500023007392883, 0.190576016902924],
  [0.499976992607117, 0.954452991485596],
  [0.366169989109039, 0.398822009563446],
  [0.393207013607025, 0.39553701877594],
  [0.410373002290726, 0.391080021858215],
  [0.194993004202843, 0.342101991176605],
  [0.388664990663528, 0.362284004688263],
  [0.365961998701096, 0.355970978736877],
  [0.343364000320435, 0.355356991291046],
  [0.318785011768341, 0.35834002494812],
  [0.301414996385574, 0.363156020641327],
  [0.058132998645306, 0.319076001644135],
  [0.301414996385574, 0.387449026107788],
  [0.499987989664078, 0.618434011936188],
  [0.415838003158569, 0.624195992946625],
  [0.445681989192963, 0.566076993942261],
  [0.465844005346298, 0.620640993118286],
  [0.49992299079895, 0.351523995399475],
  [0.288718998432159, 0.819945991039276],
  [0.335278987884521, 0.852819979190826],
  [0.440512001514435, 0.902418971061707],
  [0.128294005990028, 0.791940987110138],
  [0.408771991729736, 0.373893976211548],
  [0.455606997013092, 0.451801002025604],
  [0.499877005815506, 0.908990025520325],
  [0.375436991453171, 0.924192011356354],
  [0.11421000212431, 0.615022003650665],
  [0.448662012815475, 0.695277988910675],
  [0.4480200111866, 0.704632043838501],
  [0.447111994028091, 0.715808033943176],
  [0.444831997156143, 0.730794012546539],
  [0.430011987686157, 0.766808986663818],
  [0.406787008047104, 0.685672998428345],
  [0.400738000869751, 0.681069016456604],
  [0.392399996519089, 0.677703022956848],
  [0.367855995893478, 0.663918972015381],
  [0.247923001646996, 0.601333022117615],
  [0.452769994735718, 0.420849978923798],
  [0.43639200925827, 0.359887003898621],
  [0.416164010763168, 0.368713974952698],
  [0.413385987281799, 0.692366003990173],
  [0.228018000721931, 0.683571994304657],
  [0.468268007040024, 0.352671027183533],
  [0.411361992359161, 0.804327011108398],
  [0.499989002943039, 0.469825029373169],
  [0.479153990745544, 0.442654013633728],
  [0.499974012374878, 0.439637005329132],
  [0.432112008333206, 0.493588984012604],
  [0.499886006116867, 0.866917014122009],
  [0.49991300702095, 0.821729004383087],
  [0.456548988819122, 0.819200992584229],
  [0.344549000263214, 0.745438992977142],
  [0.37890899181366, 0.574010014533997],
  [0.374292999505997, 0.780184984207153],
  [0.319687992334366, 0.570737957954407],
  [0.357154995203018, 0.604269981384277],
  [0.295284003019333, 0.621580958366394],
  [0.447750002145767, 0.862477004528046],
  [0.410986006259918, 0.508723020553589],
  [0.31395098567009, 0.775308012962341],
  [0.354128003120422, 0.812552988529205],
  [0.324548006057739, 0.703992962837219],
  [0.189096003770828, 0.646299958229065],
  [0.279776990413666, 0.71465802192688],
  [0.1338230073452, 0.682700991630554],
  [0.336768001317978, 0.644733011722565],
  [0.429883986711502, 0.466521978378296],
  [0.455527991056442, 0.548622965812683],
  [0.437114000320435, 0.558896005153656],
  [0.467287987470627, 0.529924988746643],
  [0.414712011814117, 0.335219979286194],
  [0.37704598903656, 0.322777986526489],
  [0.344107985496521, 0.320150971412659],
  [0.312875986099243, 0.32233202457428],
  [0.283526003360748, 0.333190023899078],
  [0.241245999932289, 0.382785975933075],
  [0.102986000478268, 0.468762993812561],
  [0.267612010240555, 0.424560010433197],
  [0.297879010438919, 0.433175981044769],
  [0.333433985710144, 0.433878004550934],
  [0.366427004337311, 0.426115989685059],
  [0.396012008190155, 0.416696012020111],
  [0.420121014118195, 0.41022801399231],
  [0.007561000064015, 0.480777025222778],
  [0.432949006557465, 0.569517970085144],
  [0.458638995885849, 0.479089021682739],
  [0.473466008901596, 0.545744001865387],
  [0.476087987422943, 0.563830018043518],
  [0.468472003936768, 0.555056989192963],
  [0.433990985155106, 0.582361996173859],
  [0.483518004417419, 0.562983989715576],
  [0.482482999563217, 0.57784903049469],
  [0.42645001411438, 0.389798998832703],
  [0.438998997211456, 0.39649498462677],
  [0.450067013502121, 0.400434017181396],
  [0.289712011814117, 0.368252992630005],
  [0.276670008897781, 0.363372981548309],
  [0.517862021923065, 0.471948027610779],
  [0.710287988185883, 0.380764007568359],
  [0.526226997375488, 0.573909997940063],
  [0.895093023777008, 0.254140973091125],
  [0.634069979190826, 0.409575998783112],
  [0.661242008209229, 0.41302502155304],
  [0.688880026340485, 0.409460008144379],
  [0.725341975688934, 0.389131009578705],
  [0.606630027294159, 0.40370500087738],
  [0.654766023159027, 0.344011008739471],
  [0.629905998706818, 0.346076011657715],
  [0.680678009986877, 0.347265005111694],
  [0.702096998691559, 0.353591024875641],
  [0.75221198797226, 0.410804986953735],
  [0.602918028831482, 0.842862963676453],
  [0.719901978969574, 0.375599980354309],
  [0.893692970275879, 0.399959981441498],
  [0.790081977844238, 0.391354024410248],
  [0.643998026847839, 0.534487962722778],
  [0.528249025344849, 0.65040397644043],
  [0.525849997997284, 0.680191040039062],
  [0.560214996337891, 0.657229006290436],
  [0.585384011268616, 0.66654098033905],
  [0.549625992774963, 0.680860996246338],
  [0.57122802734375, 0.682691991329193],
  [0.624852001667023, 0.72809898853302],
  [0.513050019741058, 0.547281980514526],
  [0.51509702205658, 0.527251958847046],
  [0.742246985435486, 0.314507007598877],
  [0.598631024360657, 0.454979002475739],
  [0.570338010787964, 0.548575043678284],
  [0.578631997108459, 0.533622980117798],
  [0.723087012767792, 0.532054007053375],
  [0.516445994377136, 0.499638974666595],
  [0.662801027297974, 0.282917976379395],
  [0.70362401008606, 0.293271005153656],
  [0.830704987049103, 0.193813979625702],
  [0.552385985851288, 0.302568018436432],
  [0.607609987258911, 0.353887975215912],
  [0.645429015159607, 0.696707010269165],
  [0.932694971561432, 0.730105042457581],
  [0.557260990142822, 0.572826027870178],
  [0.542901992797852, 0.584792017936707],
  [0.6180260181427, 0.694710969924927],
  [0.607590973377228, 0.694203019142151],
  [0.722943007946014, 0.271963000297546],
  [0.577413976192474, 0.563166975975037],
  [0.614082992076874, 0.281386971473694],
  [0.616907000541687, 0.255886018276215],
  [0.668509006500244, 0.119913995265961],
  [0.770092010498047, 0.232020974159241],
  [0.635536015033722, 0.189248979091644],
  [0.77039098739624, 0.299556016921997],
  [0.826722025871277, 0.278755009174347],
  [0.527121007442474, 0.666198015213013],
  [0.553171992301941, 0.668527007102966],
  [0.577238023281097, 0.673889994621277],
  [0.554691970348358, 0.580065965652466],
  [0.611896991729736, 0.693961024284363],
  [0.59696102142334, 0.706539988517761],
  [0.596370995044708, 0.693953037261963],
  [0.539958000183105, 0.557139039039612],
  [0.568841993808746, 0.692366003990173],
  [0.547818005084991, 0.692366003990173],
  [0.52461302280426, 0.692366003990173],
  [0.534089982509613, 0.779141008853912],
  [0.527670979499817, 0.736225962638855],
  [0.526912987232208, 0.717857003211975],
  [0.526877999305725, 0.704625964164734],
  [0.526966989040375, 0.695277988910675],
  [0.572058022022247, 0.695277988910675],
  [0.573521018028259, 0.703539967536926],
  [0.57683801651001, 0.711845993995667],
  [0.581691026687622, 0.720062971115112],
  [0.609944999217987, 0.639909982681274],
  [0.986046016216278, 0.560034036636353],
  [0.5867999792099, 0.69539999961853],
  [0.590372025966644, 0.701822996139526],
  [0.531915009021759, 0.601536989212036],
  [0.577268004417419, 0.585934996604919],
  [0.536915004253387, 0.593786001205444],
  [0.627542972564697, 0.473352015018463],
  [0.665585994720459, 0.495950996875763],
  [0.588353991508484, 0.546862006187439],
  [0.757824003696442, 0.14767599105835],
  [0.709249973297119, 0.201507985591888],
  [0.672684013843536, 0.256581008434296],
  [0.600408971309662, 0.74900496006012],
  [0.55826598405838, 0.261672019958496],
  [0.570303976535797, 0.187870979309082],
  [0.588165998458862, 0.109044015407562],
  [0.711045026779175, 0.398952007293701],
  [0.781069993972778, 0.435405015945435],
  [0.587247014045715, 0.398931980133057],
  [0.742869973182678, 0.355445981025696],
  [0.572156012058258, 0.437651991844177],
  [0.55186802148819, 0.536570012569427],
  [0.821442008018494, 0.457556009292603],
  [0.752701997756958, 0.457181990146637],
  [0.71375697851181, 0.467626988887787],
  [0.66711300611496, 0.460672974586487],
  [0.631101012229919, 0.447153985500336],
  [0.6008620262146, 0.432473003864288],
  [0.523481011390686, 0.405627012252808],
  [0.810747981071472, 0.523926019668579],
  [0.771045982837677, 0.348959028720856],
  [0.509127020835876, 0.562718033790588],
  [0.595292985439301, 0.485023975372314],
  [0.980530977249146, 0.401564002037048],
  [0.573499977588654, 0.420000016689301],
  [0.602994978427887, 0.548687994480133],
  [0.733529984951019, 0.376977026462555],
  [0.560611009597778, 0.519016981124878],
  [0.967685997486115, 0.644356966018677],
  [0.580985009670258, 0.387160003185272],
  [0.537728011608124, 0.505385041236877],
  [0.760966002941132, 0.779752969741821],
  [0.801778972148895, 0.831938028335571],
  [0.892440974712372, 0.54076099395752],
  [0.816350996494293, 0.740260004997253],
  [0.865594983100891, 0.333687007427216],
  [0.614073991775513, 0.883246004581451],
  [0.508952975273132, 0.579437971115112],
  [0.617941975593567, 0.508316040039062],
  [0.825608015060425, 0.397674977779388],
  [0.681214988231659, 0.39623498916626],
  [0.656635999679565, 0.400596976280212],
  [0.603900015354156, 0.710216999053955],
  [0.81208598613739, 0.588539004325867],
  [0.56801301240921, 0.944564998149872],
  [0.681007981300354, 0.898285031318665],
  [0.733752012252808, 0.869701027870178],
  [0.633830010890961, 0.398822009563446],
  [0.606792986392975, 0.39553701877594],
  [0.589659988880157, 0.391062021255493],
  [0.805015981197357, 0.342108011245728],
  [0.611334979534149, 0.362284004688263],
  [0.634037971496582, 0.355970978736877],
  [0.656635999679565, 0.355356991291046],
  [0.681214988231659, 0.35834002494812],
  [0.698584973812103, 0.363156020641327],
  [0.941866993904114, 0.319076001644135],
  [0.698584973812103, 0.387449026107788],
  [0.584177017211914, 0.624107003211975],
  [0.554318010807037, 0.566076993942261],
  [0.534153997898102, 0.62064003944397],
  [0.711217999458313, 0.819975018501282],
  [0.664629995822906, 0.852871000766754],
  [0.559099972248077, 0.902631998062134],
  [0.871706008911133, 0.791940987110138],
  [0.591234028339386, 0.373893976211548],
  [0.544341027736664, 0.451583981513977],
  [0.624562978744507, 0.924192011356354],
  [0.88577002286911, 0.615028977394104],
  [0.551338016986847, 0.695277988910675],
  [0.551980018615723, 0.704632043838501],
  [0.552887976169586, 0.715808033943176],
  [0.555167973041534, 0.730794012546539],
  [0.569944024085999, 0.767035007476807],
  [0.593203008174896, 0.685675978660583],
  [0.599261999130249, 0.681069016456604],
  [0.607599973678589, 0.677703022956848],
  [0.631937980651855, 0.663500010967255],
  [0.752032995223999, 0.601315021514893],
  [0.547226011753082, 0.420395016670227],
  [0.563543975353241, 0.359827995300293],
  [0.583841025829315, 0.368713974952698],
  [0.586614012718201, 0.692366003990173],
  [0.771915018558502, 0.683578014373779],
  [0.531597018241882, 0.352482974529266],
  [0.588370978832245, 0.804440975189209],
  [0.52079701423645, 0.442565023899078],
  [0.567984998226166, 0.493479013442993],
  [0.543282985687256, 0.819254994392395],
  [0.655317008495331, 0.745514988899231],
  [0.621008992195129, 0.574018001556396],
  [0.625559985637665, 0.78031200170517],
  [0.680198013782501, 0.570719003677368],
  [0.64276397228241, 0.604337990283966],
  [0.704662978649139, 0.621529996395111],
  [0.552012026309967, 0.862591981887817],
  [0.589071989059448, 0.508637011051178],
  [0.685944974422455, 0.775357007980347],
  [0.645735025405884, 0.812640011310577],
  [0.675342977046967, 0.703978002071381],
  [0.810858011245728, 0.646304965019226],
  [0.72012197971344, 0.714666962623596],
  [0.866151988506317, 0.682704985141754],
  [0.663187026977539, 0.644596993923187],
  [0.570082008838654, 0.466325998306274],
  [0.544561982154846, 0.548375964164734],
  [0.562758982181549, 0.558784961700439],
  [0.531987011432648, 0.530140042304993],
  [0.585271000862122, 0.335177004337311],
  [0.622952997684479, 0.32277899980545],
  [0.655896008014679, 0.320163011550903],
  [0.687132000923157, 0.322345972061157],
  [0.716481983661652, 0.333200991153717],
  [0.758756995201111, 0.382786989212036],
  [0.897013008594513, 0.468769013881683],
  [0.732392013072968, 0.424547016620636],
  [0.70211398601532, 0.433162987232208],
  [0.66652500629425, 0.433866024017334],
  [0.633504986763, 0.426087975502014],
  [0.603875994682312, 0.416586995124817],
  [0.579657971858978, 0.409945011138916],
  [0.992439985275269, 0.480777025222778],
  [0.567192018032074, 0.569419980049133],
  [0.54136598110199, 0.478899002075195],
  [0.526564002037048, 0.546118021011353],
  [0.523913025856018, 0.563830018043518],
  [0.531529009342194, 0.555056989192963],
  [0.566035985946655, 0.582329034805298],
  [0.51631098985672, 0.563053965568542],
  [0.5174720287323, 0.577877044677734],
  [0.573594987392426, 0.389806985855103],
  [0.560697972774506, 0.395331978797913],
  [0.549755990505219, 0.399751007556915],
  [0.710287988185883, 0.368252992630005],
  [0.723330020904541, 0.363372981548309]
];
var TRI468 = [
  127,
  34,
  139,
  11,
  0,
  37,
  232,
  231,
  120,
  72,
  37,
  39,
  128,
  121,
  47,
  232,
  121,
  128,
  104,
  69,
  67,
  175,
  171,
  148,
  157,
  154,
  155,
  118,
  50,
  101,
  73,
  39,
  40,
  9,
  151,
  108,
  48,
  115,
  131,
  194,
  204,
  211,
  74,
  40,
  185,
  80,
  42,
  183,
  40,
  92,
  186,
  230,
  229,
  118,
  202,
  212,
  214,
  83,
  18,
  17,
  76,
  61,
  146,
  160,
  29,
  30,
  56,
  157,
  173,
  106,
  204,
  194,
  135,
  214,
  192,
  203,
  165,
  98,
  21,
  71,
  68,
  51,
  45,
  4,
  144,
  24,
  23,
  77,
  146,
  91,
  205,
  50,
  187,
  201,
  200,
  18,
  91,
  106,
  182,
  90,
  91,
  181,
  85,
  84,
  17,
  206,
  203,
  36,
  148,
  171,
  140,
  92,
  40,
  39,
  193,
  189,
  244,
  159,
  158,
  28,
  247,
  246,
  161,
  236,
  3,
  196,
  54,
  68,
  104,
  193,
  168,
  8,
  117,
  228,
  31,
  189,
  193,
  55,
  98,
  97,
  99,
  126,
  47,
  100,
  166,
  79,
  218,
  155,
  154,
  26,
  209,
  49,
  131,
  135,
  136,
  150,
  47,
  126,
  217,
  223,
  52,
  53,
  45,
  51,
  134,
  211,
  170,
  140,
  67,
  69,
  108,
  43,
  106,
  91,
  230,
  119,
  120,
  226,
  130,
  247,
  63,
  53,
  52,
  238,
  20,
  242,
  46,
  70,
  156,
  78,
  62,
  96,
  46,
  53,
  63,
  143,
  34,
  227,
  173,
  155,
  133,
  123,
  117,
  111,
  44,
  125,
  19,
  236,
  134,
  51,
  216,
  206,
  205,
  154,
  153,
  22,
  39,
  37,
  167,
  200,
  201,
  208,
  36,
  142,
  100,
  57,
  212,
  202,
  20,
  60,
  99,
  28,
  158,
  157,
  35,
  226,
  113,
  160,
  159,
  27,
  204,
  202,
  210,
  113,
  225,
  46,
  43,
  202,
  204,
  62,
  76,
  77,
  137,
  123,
  116,
  41,
  38,
  72,
  203,
  129,
  142,
  64,
  98,
  240,
  49,
  102,
  64,
  41,
  73,
  74,
  212,
  216,
  207,
  42,
  74,
  184,
  169,
  170,
  211,
  170,
  149,
  176,
  105,
  66,
  69,
  122,
  6,
  168,
  123,
  147,
  187,
  96,
  77,
  90,
  65,
  55,
  107,
  89,
  90,
  180,
  101,
  100,
  120,
  63,
  105,
  104,
  93,
  137,
  227,
  15,
  86,
  85,
  129,
  102,
  49,
  14,
  87,
  86,
  55,
  8,
  9,
  100,
  47,
  121,
  145,
  23,
  22,
  88,
  89,
  179,
  6,
  122,
  196,
  88,
  95,
  96,
  138,
  172,
  136,
  215,
  58,
  172,
  115,
  48,
  219,
  42,
  80,
  81,
  195,
  3,
  51,
  43,
  146,
  61,
  171,
  175,
  199,
  81,
  82,
  38,
  53,
  46,
  225,
  144,
  163,
  110,
  246,
  33,
  7,
  52,
  65,
  66,
  229,
  228,
  117,
  34,
  127,
  234,
  107,
  108,
  69,
  109,
  108,
  151,
  48,
  64,
  235,
  62,
  78,
  191,
  129,
  209,
  126,
  111,
  35,
  143,
  163,
  161,
  246,
  117,
  123,
  50,
  222,
  65,
  52,
  19,
  125,
  141,
  221,
  55,
  65,
  3,
  195,
  197,
  25,
  7,
  33,
  220,
  237,
  44,
  70,
  71,
  139,
  122,
  193,
  245,
  247,
  130,
  33,
  71,
  21,
  162,
  153,
  158,
  159,
  170,
  169,
  150,
  188,
  174,
  196,
  216,
  186,
  92,
  144,
  160,
  161,
  2,
  97,
  167,
  141,
  125,
  241,
  164,
  167,
  37,
  72,
  38,
  12,
  145,
  159,
  160,
  38,
  82,
  13,
  63,
  68,
  71,
  226,
  35,
  111,
  158,
  153,
  154,
  101,
  50,
  205,
  206,
  92,
  165,
  209,
  198,
  217,
  165,
  167,
  97,
  220,
  115,
  218,
  133,
  112,
  243,
  239,
  238,
  241,
  214,
  135,
  169,
  190,
  173,
  133,
  171,
  208,
  32,
  125,
  44,
  237,
  86,
  87,
  178,
  85,
  86,
  179,
  84,
  85,
  180,
  83,
  84,
  181,
  201,
  83,
  182,
  137,
  93,
  132,
  76,
  62,
  183,
  61,
  76,
  184,
  57,
  61,
  185,
  212,
  57,
  186,
  214,
  207,
  187,
  34,
  143,
  156,
  79,
  239,
  237,
  123,
  137,
  177,
  44,
  1,
  4,
  201,
  194,
  32,
  64,
  102,
  129,
  213,
  215,
  138,
  59,
  166,
  219,
  242,
  99,
  97,
  2,
  94,
  141,
  75,
  59,
  235,
  24,
  110,
  228,
  25,
  130,
  226,
  23,
  24,
  229,
  22,
  23,
  230,
  26,
  22,
  231,
  112,
  26,
  232,
  189,
  190,
  243,
  221,
  56,
  190,
  28,
  56,
  221,
  27,
  28,
  222,
  29,
  27,
  223,
  30,
  29,
  224,
  247,
  30,
  225,
  238,
  79,
  20,
  166,
  59,
  75,
  60,
  75,
  240,
  147,
  177,
  215,
  20,
  79,
  166,
  187,
  147,
  213,
  112,
  233,
  244,
  233,
  128,
  245,
  128,
  114,
  188,
  114,
  217,
  174,
  131,
  115,
  220,
  217,
  198,
  236,
  198,
  131,
  134,
  177,
  132,
  58,
  143,
  35,
  124,
  110,
  163,
  7,
  228,
  110,
  25,
  356,
  389,
  368,
  11,
  302,
  267,
  452,
  350,
  349,
  302,
  303,
  269,
  357,
  343,
  277,
  452,
  453,
  357,
  333,
  332,
  297,
  175,
  152,
  377,
  384,
  398,
  382,
  347,
  348,
  330,
  303,
  304,
  270,
  9,
  336,
  337,
  278,
  279,
  360,
  418,
  262,
  431,
  304,
  408,
  409,
  310,
  415,
  407,
  270,
  409,
  410,
  450,
  348,
  347,
  422,
  430,
  434,
  313,
  314,
  17,
  306,
  307,
  375,
  387,
  388,
  260,
  286,
  414,
  398,
  335,
  406,
  418,
  364,
  367,
  416,
  423,
  358,
  327,
  251,
  284,
  298,
  281,
  5,
  4,
  373,
  374,
  253,
  307,
  320,
  321,
  425,
  427,
  411,
  421,
  313,
  18,
  321,
  405,
  406,
  320,
  404,
  405,
  315,
  16,
  17,
  426,
  425,
  266,
  377,
  400,
  369,
  322,
  391,
  269,
  417,
  465,
  464,
  386,
  257,
  258,
  466,
  260,
  388,
  456,
  399,
  419,
  284,
  332,
  333,
  417,
  285,
  8,
  346,
  340,
  261,
  413,
  441,
  285,
  327,
  460,
  328,
  355,
  371,
  329,
  392,
  439,
  438,
  382,
  341,
  256,
  429,
  420,
  360,
  364,
  394,
  379,
  277,
  343,
  437,
  443,
  444,
  283,
  275,
  440,
  363,
  431,
  262,
  369,
  297,
  338,
  337,
  273,
  375,
  321,
  450,
  451,
  349,
  446,
  342,
  467,
  293,
  334,
  282,
  458,
  461,
  462,
  276,
  353,
  383,
  308,
  324,
  325,
  276,
  300,
  293,
  372,
  345,
  447,
  382,
  398,
  362,
  352,
  345,
  340,
  274,
  1,
  19,
  456,
  248,
  281,
  436,
  427,
  425,
  381,
  256,
  252,
  269,
  391,
  393,
  200,
  199,
  428,
  266,
  330,
  329,
  287,
  273,
  422,
  250,
  462,
  328,
  258,
  286,
  384,
  265,
  353,
  342,
  387,
  259,
  257,
  424,
  431,
  430,
  342,
  353,
  276,
  273,
  335,
  424,
  292,
  325,
  307,
  366,
  447,
  345,
  271,
  303,
  302,
  423,
  266,
  371,
  294,
  455,
  460,
  279,
  278,
  294,
  271,
  272,
  304,
  432,
  434,
  427,
  272,
  407,
  408,
  394,
  430,
  431,
  395,
  369,
  400,
  334,
  333,
  299,
  351,
  417,
  168,
  352,
  280,
  411,
  325,
  319,
  320,
  295,
  296,
  336,
  319,
  403,
  404,
  330,
  348,
  349,
  293,
  298,
  333,
  323,
  454,
  447,
  15,
  16,
  315,
  358,
  429,
  279,
  14,
  15,
  316,
  285,
  336,
  9,
  329,
  349,
  350,
  374,
  380,
  252,
  318,
  402,
  403,
  6,
  197,
  419,
  318,
  319,
  325,
  367,
  364,
  365,
  435,
  367,
  397,
  344,
  438,
  439,
  272,
  271,
  311,
  195,
  5,
  281,
  273,
  287,
  291,
  396,
  428,
  199,
  311,
  271,
  268,
  283,
  444,
  445,
  373,
  254,
  339,
  263,
  466,
  249,
  282,
  334,
  296,
  449,
  347,
  346,
  264,
  447,
  454,
  336,
  296,
  299,
  338,
  10,
  151,
  278,
  439,
  455,
  292,
  407,
  415,
  358,
  371,
  355,
  340,
  345,
  372,
  390,
  249,
  466,
  346,
  347,
  280,
  442,
  443,
  282,
  19,
  94,
  370,
  441,
  442,
  295,
  248,
  419,
  197,
  263,
  255,
  359,
  440,
  275,
  274,
  300,
  383,
  368,
  351,
  412,
  465,
  263,
  467,
  466,
  301,
  368,
  389,
  380,
  374,
  386,
  395,
  378,
  379,
  412,
  351,
  419,
  436,
  426,
  322,
  373,
  390,
  388,
  2,
  164,
  393,
  370,
  462,
  461,
  164,
  0,
  267,
  302,
  11,
  12,
  374,
  373,
  387,
  268,
  12,
  13,
  293,
  300,
  301,
  446,
  261,
  340,
  385,
  384,
  381,
  330,
  266,
  425,
  426,
  423,
  391,
  429,
  355,
  437,
  391,
  327,
  326,
  440,
  457,
  438,
  341,
  382,
  362,
  459,
  457,
  461,
  434,
  430,
  394,
  414,
  463,
  362,
  396,
  369,
  262,
  354,
  461,
  457,
  316,
  403,
  402,
  315,
  404,
  403,
  314,
  405,
  404,
  313,
  406,
  405,
  421,
  418,
  406,
  366,
  401,
  361,
  306,
  408,
  407,
  291,
  409,
  408,
  287,
  410,
  409,
  432,
  436,
  410,
  434,
  416,
  411,
  264,
  368,
  383,
  309,
  438,
  457,
  352,
  376,
  401,
  274,
  275,
  4,
  421,
  428,
  262,
  294,
  327,
  358,
  433,
  416,
  367,
  289,
  455,
  439,
  462,
  370,
  326,
  2,
  326,
  370,
  305,
  460,
  455,
  254,
  449,
  448,
  255,
  261,
  446,
  253,
  450,
  449,
  252,
  451,
  450,
  256,
  452,
  451,
  341,
  453,
  452,
  413,
  464,
  463,
  441,
  413,
  414,
  258,
  442,
  441,
  257,
  443,
  442,
  259,
  444,
  443,
  260,
  445,
  444,
  467,
  342,
  445,
  459,
  458,
  250,
  289,
  392,
  290,
  290,
  328,
  460,
  376,
  433,
  435,
  250,
  290,
  392,
  411,
  416,
  433,
  341,
  463,
  464,
  453,
  464,
  465,
  357,
  465,
  412,
  343,
  412,
  399,
  360,
  363,
  440,
  437,
  399,
  456,
  420,
  456,
  363,
  401,
  435,
  288,
  372,
  383,
  353,
  339,
  255,
  249,
  448,
  261,
  255,
  133,
  243,
  190,
  133,
  155,
  112,
  33,
  246,
  247,
  33,
  130,
  25,
  398,
  384,
  286,
  362,
  398,
  414,
  362,
  463,
  341,
  263,
  359,
  467,
  263,
  249,
  255,
  466,
  467,
  260,
  75,
  60,
  166,
  238,
  239,
  79,
  162,
  127,
  139,
  72,
  11,
  37,
  121,
  232,
  120,
  73,
  72,
  39,
  114,
  128,
  47,
  233,
  232,
  128,
  103,
  104,
  67,
  152,
  175,
  148,
  173,
  157,
  155,
  119,
  118,
  101,
  74,
  73,
  40,
  107,
  9,
  108,
  49,
  48,
  131,
  32,
  194,
  211,
  184,
  74,
  185,
  191,
  80,
  183,
  185,
  40,
  186,
  119,
  230,
  118,
  210,
  202,
  214,
  84,
  83,
  17,
  77,
  76,
  146,
  161,
  160,
  30,
  190,
  56,
  173,
  182,
  106,
  194,
  138,
  135,
  192,
  129,
  203,
  98,
  54,
  21,
  68,
  5,
  51,
  4,
  145,
  144,
  23,
  90,
  77,
  91,
  207,
  205,
  187,
  83,
  201,
  18,
  181,
  91,
  182,
  180,
  90,
  181,
  16,
  85,
  17,
  205,
  206,
  36,
  176,
  148,
  140,
  165,
  92,
  39,
  245,
  193,
  244,
  27,
  159,
  28,
  30,
  247,
  161,
  174,
  236,
  196,
  103,
  54,
  104,
  55,
  193,
  8,
  111,
  117,
  31,
  221,
  189,
  55,
  240,
  98,
  99,
  142,
  126,
  100,
  219,
  166,
  218,
  112,
  155,
  26,
  198,
  209,
  131,
  169,
  135,
  150,
  114,
  47,
  217,
  224,
  223,
  53,
  220,
  45,
  134,
  32,
  211,
  140,
  109,
  67,
  108,
  146,
  43,
  91,
  231,
  230,
  120,
  113,
  226,
  247,
  105,
  63,
  52,
  241,
  238,
  242,
  124,
  46,
  156,
  95,
  78,
  96,
  70,
  46,
  63,
  116,
  143,
  227,
  116,
  123,
  111,
  1,
  44,
  19,
  3,
  236,
  51,
  207,
  216,
  205,
  26,
  154,
  22,
  165,
  39,
  167,
  199,
  200,
  208,
  101,
  36,
  100,
  43,
  57,
  202,
  242,
  20,
  99,
  56,
  28,
  157,
  124,
  35,
  113,
  29,
  160,
  27,
  211,
  204,
  210,
  124,
  113,
  46,
  106,
  43,
  204,
  96,
  62,
  77,
  227,
  137,
  116,
  73,
  41,
  72,
  36,
  203,
  142,
  235,
  64,
  240,
  48,
  49,
  64,
  42,
  41,
  74,
  214,
  212,
  207,
  183,
  42,
  184,
  210,
  169,
  211,
  140,
  170,
  176,
  104,
  105,
  69,
  193,
  122,
  168,
  50,
  123,
  187,
  89,
  96,
  90,
  66,
  65,
  107,
  179,
  89,
  180,
  119,
  101,
  120,
  68,
  63,
  104,
  234,
  93,
  227,
  16,
  15,
  85,
  209,
  129,
  49,
  15,
  14,
  86,
  107,
  55,
  9,
  120,
  100,
  121,
  153,
  145,
  22,
  178,
  88,
  179,
  197,
  6,
  196,
  89,
  88,
  96,
  135,
  138,
  136,
  138,
  215,
  172,
  218,
  115,
  219,
  41,
  42,
  81,
  5,
  195,
  51,
  57,
  43,
  61,
  208,
  171,
  199,
  41,
  81,
  38,
  224,
  53,
  225,
  24,
  144,
  110,
  105,
  52,
  66,
  118,
  229,
  117,
  227,
  34,
  234,
  66,
  107,
  69,
  10,
  109,
  151,
  219,
  48,
  235,
  183,
  62,
  191,
  142,
  129,
  126,
  116,
  111,
  143,
  7,
  163,
  246,
  118,
  117,
  50,
  223,
  222,
  52,
  94,
  19,
  141,
  222,
  221,
  65,
  196,
  3,
  197,
  45,
  220,
  44,
  156,
  70,
  139,
  188,
  122,
  245,
  139,
  71,
  162,
  145,
  153,
  159,
  149,
  170,
  150,
  122,
  188,
  196,
  206,
  216,
  92,
  163,
  144,
  161,
  164,
  2,
  167,
  242,
  141,
  241,
  0,
  164,
  37,
  11,
  72,
  12,
  144,
  145,
  160,
  12,
  38,
  13,
  70,
  63,
  71,
  31,
  226,
  111,
  157,
  158,
  154,
  36,
  101,
  205,
  203,
  206,
  165,
  126,
  209,
  217,
  98,
  165,
  97,
  237,
  220,
  218,
  237,
  239,
  241,
  210,
  214,
  169,
  140,
  171,
  32,
  241,
  125,
  237,
  179,
  86,
  178,
  180,
  85,
  179,
  181,
  84,
  180,
  182,
  83,
  181,
  194,
  201,
  182,
  177,
  137,
  132,
  184,
  76,
  183,
  185,
  61,
  184,
  186,
  57,
  185,
  216,
  212,
  186,
  192,
  214,
  187,
  139,
  34,
  156,
  218,
  79,
  237,
  147,
  123,
  177,
  45,
  44,
  4,
  208,
  201,
  32,
  98,
  64,
  129,
  192,
  213,
  138,
  235,
  59,
  219,
  141,
  242,
  97,
  97,
  2,
  141,
  240,
  75,
  235,
  229,
  24,
  228,
  31,
  25,
  226,
  230,
  23,
  229,
  231,
  22,
  230,
  232,
  26,
  231,
  233,
  112,
  232,
  244,
  189,
  243,
  189,
  221,
  190,
  222,
  28,
  221,
  223,
  27,
  222,
  224,
  29,
  223,
  225,
  30,
  224,
  113,
  247,
  225,
  99,
  60,
  240,
  213,
  147,
  215,
  60,
  20,
  166,
  192,
  187,
  213,
  243,
  112,
  244,
  244,
  233,
  245,
  245,
  128,
  188,
  188,
  114,
  174,
  134,
  131,
  220,
  174,
  217,
  236,
  236,
  198,
  134,
  215,
  177,
  58,
  156,
  143,
  124,
  25,
  110,
  7,
  31,
  228,
  25,
  264,
  356,
  368,
  0,
  11,
  267,
  451,
  452,
  349,
  267,
  302,
  269,
  350,
  357,
  277,
  350,
  452,
  357,
  299,
  333,
  297,
  396,
  175,
  377,
  381,
  384,
  382,
  280,
  347,
  330,
  269,
  303,
  270,
  151,
  9,
  337,
  344,
  278,
  360,
  424,
  418,
  431,
  270,
  304,
  409,
  272,
  310,
  407,
  322,
  270,
  410,
  449,
  450,
  347,
  432,
  422,
  434,
  18,
  313,
  17,
  291,
  306,
  375,
  259,
  387,
  260,
  424,
  335,
  418,
  434,
  364,
  416,
  391,
  423,
  327,
  301,
  251,
  298,
  275,
  281,
  4,
  254,
  373,
  253,
  375,
  307,
  321,
  280,
  425,
  411,
  200,
  421,
  18,
  335,
  321,
  406,
  321,
  320,
  405,
  314,
  315,
  17,
  423,
  426,
  266,
  396,
  377,
  369,
  270,
  322,
  269,
  413,
  417,
  464,
  385,
  386,
  258,
  248,
  456,
  419,
  298,
  284,
  333,
  168,
  417,
  8,
  448,
  346,
  261,
  417,
  413,
  285,
  326,
  327,
  328,
  277,
  355,
  329,
  309,
  392,
  438,
  381,
  382,
  256,
  279,
  429,
  360,
  365,
  364,
  379,
  355,
  277,
  437,
  282,
  443,
  283,
  281,
  275,
  363,
  395,
  431,
  369,
  299,
  297,
  337,
  335,
  273,
  321,
  348,
  450,
  349,
  359,
  446,
  467,
  283,
  293,
  282,
  250,
  458,
  462,
  300,
  276,
  383,
  292,
  308,
  325,
  283,
  276,
  293,
  264,
  372,
  447,
  346,
  352,
  340,
  354,
  274,
  19,
  363,
  456,
  281,
  426,
  436,
  425,
  380,
  381,
  252,
  267,
  269,
  393,
  421,
  200,
  428,
  371,
  266,
  329,
  432,
  287,
  422,
  290,
  250,
  328,
  385,
  258,
  384,
  446,
  265,
  342,
  386,
  387,
  257,
  422,
  424,
  430,
  445,
  342,
  276,
  422,
  273,
  424,
  306,
  292,
  307,
  352,
  366,
  345,
  268,
  271,
  302,
  358,
  423,
  371,
  327,
  294,
  460,
  331,
  279,
  294,
  303,
  271,
  304,
  436,
  432,
  427,
  304,
  272,
  408,
  395,
  394,
  431,
  378,
  395,
  400,
  296,
  334,
  299,
  6,
  351,
  168,
  376,
  352,
  411,
  307,
  325,
  320,
  285,
  295,
  336,
  320,
  319,
  404,
  329,
  330,
  349,
  334,
  293,
  333,
  366,
  323,
  447,
  316,
  15,
  315,
  331,
  358,
  279,
  317,
  14,
  316,
  8,
  285,
  9,
  277,
  329,
  350,
  253,
  374,
  252,
  319,
  318,
  403,
  351,
  6,
  419,
  324,
  318,
  325,
  397,
  367,
  365,
  288,
  435,
  397,
  278,
  344,
  439,
  310,
  272,
  311,
  248,
  195,
  281,
  375,
  273,
  291,
  175,
  396,
  199,
  312,
  311,
  268,
  276,
  283,
  445,
  390,
  373,
  339,
  295,
  282,
  296,
  448,
  449,
  346,
  356,
  264,
  454,
  337,
  336,
  299,
  337,
  338,
  151,
  294,
  278,
  455,
  308,
  292,
  415,
  429,
  358,
  355,
  265,
  340,
  372,
  388,
  390,
  466,
  352,
  346,
  280,
  295,
  442,
  282,
  354,
  19,
  370,
  285,
  441,
  295,
  195,
  248,
  197,
  457,
  440,
  274,
  301,
  300,
  368,
  417,
  351,
  465,
  251,
  301,
  389,
  385,
  380,
  386,
  394,
  395,
  379,
  399,
  412,
  419,
  410,
  436,
  322,
  387,
  373,
  388,
  326,
  2,
  393,
  354,
  370,
  461,
  393,
  164,
  267,
  268,
  302,
  12,
  386,
  374,
  387,
  312,
  268,
  13,
  298,
  293,
  301,
  265,
  446,
  340,
  380,
  385,
  381,
  280,
  330,
  425,
  322,
  426,
  391,
  420,
  429,
  437,
  393,
  391,
  326,
  344,
  440,
  438,
  458,
  459,
  461,
  364,
  434,
  394,
  428,
  396,
  262,
  274,
  354,
  457,
  317,
  316,
  402,
  316,
  315,
  403,
  315,
  314,
  404,
  314,
  313,
  405,
  313,
  421,
  406,
  323,
  366,
  361,
  292,
  306,
  407,
  306,
  291,
  408,
  291,
  287,
  409,
  287,
  432,
  410,
  427,
  434,
  411,
  372,
  264,
  383,
  459,
  309,
  457,
  366,
  352,
  401,
  1,
  274,
  4,
  418,
  421,
  262,
  331,
  294,
  358,
  435,
  433,
  367,
  392,
  289,
  439,
  328,
  462,
  326,
  94,
  2,
  370,
  289,
  305,
  455,
  339,
  254,
  448,
  359,
  255,
  446,
  254,
  253,
  449,
  253,
  252,
  450,
  252,
  256,
  451,
  256,
  341,
  452,
  414,
  413,
  463,
  286,
  441,
  414,
  286,
  258,
  441,
  258,
  257,
  442,
  257,
  259,
  443,
  259,
  260,
  444,
  260,
  467,
  445,
  309,
  459,
  250,
  305,
  289,
  290,
  305,
  290,
  460,
  401,
  376,
  435,
  309,
  250,
  392,
  376,
  411,
  433,
  453,
  341,
  464,
  357,
  453,
  465,
  343,
  357,
  412,
  437,
  343,
  399,
  344,
  360,
  440,
  420,
  437,
  456,
  360,
  420,
  363,
  361,
  401,
  288,
  265,
  372,
  353,
  390,
  339,
  249,
  339,
  448,
  255
];
var VTX68 = [
  127,
  234,
  132,
  58,
  172,
  150,
  149,
  148,
  152,
  377,
  378,
  379,
  397,
  288,
  361,
  454,
  356,
  70,
  63,
  105,
  66,
  107,
  336,
  296,
  334,
  293,
  300,
  168,
  6,
  195,
  4,
  98,
  97,
  2,
  326,
  327,
  33,
  160,
  158,
  133,
  153,
  144,
  362,
  385,
  387,
  263,
  373,
  380,
  57,
  40,
  37,
  0,
  267,
  270,
  287,
  321,
  314,
  17,
  84,
  91,
  78,
  81,
  13,
  311,
  308,
  402,
  14,
  178
];
var VTX33 = [33, 133, 362, 263, 1, 62, 308, 159, 145, 386, 374, 6, 102, 331, 2, 13, 14, 70, 105, 107, 336, 334, 300, 54, 10, 284, 50, 280, 234, 454, 58, 288, 152];
var VTX7 = [33, 133, 362, 263, 1, 78, 308];
var UV68 = VTX68.map((x) => UV468[x]);
var UV33 = VTX33.map((x) => UV468[x]);
var UV7 = VTX7.map((x) => UV468[x]);
function connectionsToIndices(connections) {
  const indices = connections.map((connection) => connection[0]);
  indices.push(connections[connections.length - 1][1]);
  return indices;
}
var pairsLips = [
  [61, 146],
  [146, 91],
  [91, 181],
  [181, 84],
  [84, 17],
  [17, 314],
  [314, 405],
  [405, 321],
  [321, 375],
  [375, 291],
  [61, 185],
  [185, 40],
  [40, 39],
  [39, 37],
  [37, 0],
  [0, 267],
  [267, 269],
  [269, 270],
  [270, 409],
  [409, 291],
  [78, 95],
  [95, 88],
  [88, 178],
  [178, 87],
  [87, 14],
  [14, 317],
  [317, 402],
  [402, 318],
  [318, 324],
  [324, 308],
  [78, 191],
  [191, 80],
  [80, 81],
  [81, 82],
  [82, 13],
  [13, 312],
  [312, 311],
  [311, 310],
  [310, 415],
  [415, 308]
];
var pairsLeftEye = [[263, 249], [249, 390], [390, 373], [373, 374], [374, 380], [380, 381], [381, 382], [382, 362], [263, 466], [466, 388], [388, 387], [387, 386], [386, 385], [385, 384], [384, 398], [398, 362]];
var pairsLeftEyebrow = [[276, 283], [283, 282], [282, 295], [295, 285], [300, 293], [293, 334], [334, 296], [296, 336]];
var pairsLeftIris = [[474, 475], [475, 476], [476, 477], [477, 474]];
var pairsRightEye = [[33, 7], [7, 163], [163, 144], [144, 145], [145, 153], [153, 154], [154, 155], [155, 133], [33, 246], [246, 161], [161, 160], [160, 159], [159, 158], [158, 157], [157, 173], [173, 133]];
var pairsRightEyebrow = [[46, 53], [53, 52], [52, 65], [65, 55], [70, 63], [63, 105], [105, 66], [66, 107]];
var pairsRightIris = [[469, 470], [470, 471], [471, 472], [472, 469]];
var pairsFaceContour = [
  [10, 338],
  [338, 297],
  [297, 332],
  [332, 284],
  [284, 251],
  [251, 389],
  [389, 356],
  [356, 454],
  [454, 323],
  [323, 361],
  [361, 288],
  [288, 397],
  [397, 365],
  [365, 379],
  [379, 378],
  [378, 400],
  [400, 377],
  [377, 152],
  [152, 148],
  [148, 176],
  [176, 149],
  [149, 150],
  [150, 136],
  [136, 172],
  [172, 58],
  [58, 132],
  [132, 93],
  [93, 234],
  [234, 127],
  [127, 162],
  [162, 21],
  [21, 54],
  [54, 103],
  [103, 67],
  [67, 109],
  [109, 10]
];
var contourKeypoints = {
  lips: connectionsToIndices(pairsLips),
  leftEye: connectionsToIndices(pairsLeftEye),
  leftEyebrow: connectionsToIndices(pairsLeftEyebrow),
  leftIris: connectionsToIndices(pairsLeftIris),
  rightEye: connectionsToIndices(pairsRightEye),
  rightEyebrow: connectionsToIndices(pairsRightEyebrow),
  rightIris: connectionsToIndices(pairsRightIris),
  faceOval: connectionsToIndices(pairsFaceContour)
};

// src/face/facemeshutil.ts
var getBoxSize = (box) => [Math.abs(box.endPoint[0] - box.startPoint[0]), Math.abs(box.endPoint[1] - box.startPoint[1])];
var getBoxCenter = (box) => [box.startPoint[0] + (box.endPoint[0] - box.startPoint[0]) / 2, box.startPoint[1] + (box.endPoint[1] - box.startPoint[1]) / 2, 1];
var clampBox = (box, input) => box ? [
  Math.trunc(Math.max(0, box.startPoint[0])),
  Math.trunc(Math.max(0, box.startPoint[1])),
  Math.trunc(Math.min(input.shape[2] || 0, box.endPoint[0]) - Math.max(0, box.startPoint[0])),
  Math.trunc(Math.min(input.shape[1] || 0, box.endPoint[1]) - Math.max(0, box.startPoint[1]))
] : [0, 0, 0, 0];
var getRawBox = (box, input) => box ? [
  box.startPoint[0] / (input.shape[2] || 0),
  box.startPoint[1] / (input.shape[1] || 0),
  (box.endPoint[0] - box.startPoint[0]) / (input.shape[2] || 0),
  (box.endPoint[1] - box.startPoint[1]) / (input.shape[1] || 0)
] : [0, 0, 0, 0];
var scaleBoxCoordinates = (box, factor) => {
  const startPoint = [box.startPoint[0] * factor[0], box.startPoint[1] * factor[1]];
  const endPoint = [box.endPoint[0] * factor[0], box.endPoint[1] * factor[1]];
  return { startPoint, endPoint, landmarks: box.landmarks, confidence: box.confidence };
};
var cutAndResize = (box, image, cropSize) => {
  const h = image.shape[1];
  const w10 = image.shape[2];
  const cutBox = [box.startPoint[1] / h, box.startPoint[0] / w10, box.endPoint[1] / h, box.endPoint[0] / w10];
  const crop = jn.cropAndResize(image, [cutBox], [0], cropSize);
  const norm = xe(crop, constants.tf255);
  De(crop);
  return norm;
};
var enlargeBox = (box, factor) => {
  const center = getBoxCenter(box);
  const size2 = getBoxSize(box);
  const halfSize = [factor * size2[0] / 2, factor * size2[1] / 2];
  return { startPoint: [center[0] - halfSize[0], center[1] - halfSize[1]], endPoint: [center[0] + halfSize[0], center[1] + halfSize[1]], landmarks: box.landmarks, confidence: box.confidence };
};
var squarifyBox = (box) => {
  const centers = getBoxCenter(box);
  const size2 = getBoxSize(box);
  const halfSize = Math.max(...size2) / 2;
  return { startPoint: [Math.round(centers[0] - halfSize), Math.round(centers[1] - halfSize)], endPoint: [Math.round(centers[0] + halfSize), Math.round(centers[1] + halfSize)], landmarks: box.landmarks, confidence: box.confidence };
};
var calculateLandmarksBoundingBox = (landmarks) => {
  const x = landmarks.map((d) => d[0]);
  const y = landmarks.map((d) => d[1]);
  return { startPoint: [Math.min(...x), Math.min(...y)], endPoint: [Math.max(...x), Math.max(...y)], landmarks };
};
var fixedRotationMatrix = [[1, 0, 0], [0, 1, 0], [0, 0, 1]];
var normalizeRadians = (angle) => angle - 2 * Math.PI * Math.floor((angle + Math.PI) / (2 * Math.PI));
var computeRotation = (point1, point2) => normalizeRadians(Math.PI / 2 - Math.atan2(-(point2[1] - point1[1]), point2[0] - point1[0]));
var buildTranslationMatrix = (x, y) => [[1, 0, x], [0, 1, y], [0, 0, 1]];
var dot = (v12, v22) => {
  let product = 0;
  for (let i = 0; i < v12.length; i++)
    product += v12[i] * v22[i];
  return product;
};
var getColumnFrom2DArr = (arr, columnIndex) => {
  const column = [];
  for (let i = 0; i < arr.length; i++)
    column.push(arr[i][columnIndex]);
  return column;
};
var multiplyTransformMatrices = (mat1, mat2) => {
  const product = [];
  const size2 = mat1.length;
  for (let row = 0; row < size2; row++) {
    product.push([]);
    for (let col = 0; col < size2; col++)
      product[row].push(dot(mat1[row], getColumnFrom2DArr(mat2, col)));
  }
  return product;
};
var buildRotationMatrix = (rotation, center) => {
  const cosA = Math.cos(rotation);
  const sinA = Math.sin(rotation);
  const rotationMatrix = [[cosA, -sinA, 0], [sinA, cosA, 0], [0, 0, 1]];
  const translationMatrix = buildTranslationMatrix(center[0], center[1]);
  const translationTimesRotation = multiplyTransformMatrices(translationMatrix, rotationMatrix);
  const negativeTranslationMatrix = buildTranslationMatrix(-center[0], -center[1]);
  return multiplyTransformMatrices(translationTimesRotation, negativeTranslationMatrix);
};
var invertTransformMatrix = (matrix) => {
  const rotationComponent = [[matrix[0][0], matrix[1][0]], [matrix[0][1], matrix[1][1]]];
  const translationComponent = [matrix[0][2], matrix[1][2]];
  const invertedTranslation = [-dot(rotationComponent[0], translationComponent), -dot(rotationComponent[1], translationComponent)];
  return [rotationComponent[0].concat(invertedTranslation[0]), rotationComponent[1].concat(invertedTranslation[1]), [0, 0, 1]];
};
var rotatePoint = (homogeneousCoordinate, rotationMatrix) => [dot(homogeneousCoordinate, rotationMatrix[0]), dot(homogeneousCoordinate, rotationMatrix[1])];
function generateAnchors(inputSize10) {
  const spec = { strides: [inputSize10 / 16, inputSize10 / 8], anchors: [2, 6] };
  const anchors3 = [];
  for (let i = 0; i < spec.strides.length; i++) {
    const stride = spec.strides[i];
    const gridRows = Math.floor((inputSize10 + stride - 1) / stride);
    const gridCols = Math.floor((inputSize10 + stride - 1) / stride);
    const anchorsNum = spec.anchors[i];
    for (let gridY = 0; gridY < gridRows; gridY++) {
      const anchorY = stride * (gridY + 0.5);
      for (let gridX = 0; gridX < gridCols; gridX++) {
        const anchorX = stride * (gridX + 0.5);
        for (let n = 0; n < anchorsNum; n++)
          anchors3.push([anchorX, anchorY]);
      }
    }
  }
  return anchors3;
}
function transformRawCoords(coordsRaw, box, angle, rotationMatrix, inputSize10) {
  const boxSize = getBoxSize(box);
  const coordsScaled = coordsRaw.map((coord) => [
    boxSize[0] / inputSize10 * (coord[0] - inputSize10 / 2),
    boxSize[1] / inputSize10 * (coord[1] - inputSize10 / 2),
    coord[2] || 0
  ]);
  const largeAngle = angle && angle !== 0 && Math.abs(angle) > 0.2;
  const coordsRotationMatrix = largeAngle ? buildRotationMatrix(angle, [0, 0]) : fixedRotationMatrix;
  const coordsRotated = largeAngle ? coordsScaled.map((coord) => [...rotatePoint(coord, coordsRotationMatrix), coord[2]]) : coordsScaled;
  const inverseRotationMatrix = largeAngle ? invertTransformMatrix(rotationMatrix) : fixedRotationMatrix;
  const boxCenter = getBoxCenter(box);
  const offsets = [dot(boxCenter, inverseRotationMatrix[0]), dot(boxCenter, inverseRotationMatrix[1])];
  return coordsRotated.map((coord) => [
    Math.trunc(coord[0] + offsets[0]),
    Math.trunc(coord[1] + offsets[1]),
    Math.trunc(coord[2] || 0)
  ]);
}
function correctFaceRotation(rotate, box, input, inputSize10) {
  const symmetryLine = box.landmarks.length >= meshLandmarks.count ? meshLandmarks.symmetryLine : blazeFaceLandmarks.symmetryLine;
  let angle = 0;
  let rotationMatrix = fixedRotationMatrix;
  let face4;
  if (rotate && env.kernels.includes("rotatewithoffset")) {
    angle = computeRotation(box.landmarks[symmetryLine[0]], box.landmarks[symmetryLine[1]]);
    const largeAngle = angle && angle !== 0 && Math.abs(angle) > 0.2;
    if (largeAngle) {
      const center = getBoxCenter(box);
      const centerRaw = [center[0] / input.shape[2], center[1] / input.shape[1]];
      const rotated = jn.rotateWithOffset(input, angle, 0, centerRaw);
      rotationMatrix = buildRotationMatrix(-angle, center);
      face4 = cutAndResize(box, rotated, [inputSize10, inputSize10]);
      De(rotated);
    } else {
      face4 = cutAndResize(box, input, [inputSize10, inputSize10]);
    }
  } else {
    face4 = cutAndResize(box, input, [inputSize10, inputSize10]);
  }
  return [angle, rotationMatrix, face4];
}
var findFaceCenter = (mesh) => {
  const x = mesh.map((m) => m[0]);
  const y = mesh.map((m) => m[1]);
  return [Math.min(...x) + (Math.max(...x) - Math.min(...x)) / 2, Math.min(...y) + (Math.max(...y) - Math.min(...y)) / 2];
};
var calculateFaceBox = (mesh, previousBox) => {
  const center = findFaceCenter(mesh);
  const boxSize = getBoxSize(previousBox);
  const calculatedBox = {
    startPoint: [center[0] - boxSize[0] / 2, center[1] - boxSize[1] / 2],
    endPoint: [center[0] + boxSize[0] / 2, center[1] + boxSize[1] / 2]
  };
  return calculatedBox;
};

// src/face/blazeface.ts
var keypointsCount = 6;
var faceBoxScaleFactor = 1.4;
var model5;
var anchors = null;
var inputSize = 0;
var inputSizeT = null;
var size = () => inputSize;
async function load5(config3) {
  var _a2;
  if (env.initial)
    model5 = null;
  if (!model5)
    model5 = await loadModel((_a2 = config3.face.detector) == null ? void 0 : _a2.modelPath);
  else if (config3.debug)
    log("cached model:", model5["modelUrl"]);
  inputSize = model5.inputs[0].shape ? model5.inputs[0].shape[2] : 0;
  inputSizeT = we(inputSize, "int32");
  anchors = Zi(generateAnchors(inputSize));
  return model5;
}
function decodeBounds(boxOutputs) {
  const t = {};
  t.boxStarts = qe(boxOutputs, [0, 1], [-1, 2]);
  t.centers = ie(t.boxStarts, anchors);
  t.boxSizes = qe(boxOutputs, [0, 3], [-1, 2]);
  t.boxSizesNormalized = xe(t.boxSizes, inputSizeT);
  t.centersNormalized = xe(t.centers, inputSizeT);
  t.halfBoxSize = xe(t.boxSizesNormalized, constants.tf2);
  t.starts = ge(t.centersNormalized, t.halfBoxSize);
  t.ends = ie(t.centersNormalized, t.halfBoxSize);
  t.startNormalized = V(t.starts, inputSizeT);
  t.endNormalized = V(t.ends, inputSizeT);
  const boxes = QE([t.startNormalized, t.endNormalized], 1);
  Object.keys(t).forEach((tensor) => De(t[tensor]));
  return boxes;
}
async function getBoxes(inputImage, config3) {
  var _a2, _b2, _c, _d2;
  if (!inputImage || inputImage["isDisposedInternal"] || inputImage.shape.length !== 4 || inputImage.shape[1] < 1 || inputImage.shape[2] < 1)
    return [];
  const t = {};
  t.resized = jn.resizeBilinear(inputImage, [inputSize, inputSize]);
  t.div = xe(t.resized, constants.tf127);
  t.normalized = ge(t.div, constants.tf05);
  const res = model5 == null ? void 0 : model5.execute(t.normalized);
  if (Array.isArray(res) && res.length > 2) {
    const sorted = res.sort((a, b) => a.size - b.size);
    t.concat384 = Ot([sorted[0], sorted[2]], 2);
    t.concat512 = Ot([sorted[1], sorted[3]], 2);
    t.concat = Ot([t.concat512, t.concat384], 1);
    t.batch = mr(t.concat, 0);
  } else if (Array.isArray(res)) {
    t.batch = mr(res[0]);
  } else {
    t.batch = mr(res);
  }
  De(res);
  t.boxes = decodeBounds(t.batch);
  t.logits = qe(t.batch, [0, 0], [-1, 1]);
  t.sigmoid = Hs(t.logits);
  t.scores = mr(t.sigmoid);
  t.nms = await jn.nonMaxSuppressionAsync(t.boxes, t.scores, ((_a2 = config3.face.detector) == null ? void 0 : _a2.maxDetected) || 0, ((_b2 = config3.face.detector) == null ? void 0 : _b2.iouThreshold) || 0, ((_c = config3.face.detector) == null ? void 0 : _c.minConfidence) || 0);
  const nms = await t.nms.array();
  const boxes = [];
  const scores = await t.scores.data();
  for (let i = 0; i < nms.length; i++) {
    const confidence = scores[nms[i]];
    if (confidence > (((_d2 = config3.face.detector) == null ? void 0 : _d2.minConfidence) || 0)) {
      const b = {};
      b.bbox = qe(t.boxes, [nms[i], 0], [1, -1]);
      b.slice = qe(t.batch, [nms[i], keypointsCount - 1], [1, -1]);
      b.squeeze = mr(b.slice);
      b.landmarks = U(b.squeeze, [keypointsCount, -1]);
      const points = await b.bbox.data();
      const rawBox = {
        startPoint: [points[0], points[1]],
        endPoint: [points[2], points[3]],
        landmarks: await b.landmarks.array(),
        confidence
      };
      const scaledBox = scaleBoxCoordinates(rawBox, [(inputImage.shape[2] || 0) / inputSize, (inputImage.shape[1] || 0) / inputSize]);
      const enlargedBox = enlargeBox(scaledBox, config3.face["scale"] || faceBoxScaleFactor);
      const squaredBox = squarifyBox(enlargedBox);
      boxes.push(squaredBox);
      Object.keys(b).forEach((tensor) => De(b[tensor]));
    }
  }
  Object.keys(t).forEach((tensor) => De(t[tensor]));
  return boxes;
}

// src/body/blazeposecoords.ts
var blazeposecoords_exports = {};
__export(blazeposecoords_exports, {
  connected: () => connected,
  kpt: () => kpt
});
var kpt = [
  "nose",
  "leftEyeInside",
  "leftEye",
  "leftEyeOutside",
  "rightEyeInside",
  "rightEye",
  "rightEyeOutside",
  "leftEar",
  "rightEar",
  "leftMouth",
  "rightMouth",
  "leftShoulder",
  "rightShoulder",
  "leftElbow",
  "rightElbow",
  "leftWrist",
  "rightWrist",
  "leftPinky",
  "rightPinky",
  "leftIndex",
  "rightIndex",
  "leftThumb",
  "rightThumb",
  "leftHip",
  "rightHip",
  "leftKnee",
  "rightKnee",
  "leftAnkle",
  "rightAnkle",
  "leftHeel",
  "rightHeel",
  "leftFoot",
  "rightFoot",
  "bodyCenter",
  "bodyTop",
  "leftPalm",
  "leftHand",
  "rightPalm",
  "rightHand"
];
var connected = {
  shoulders: ["leftShoulder", "rightShoulder"],
  hips: ["rightHip", "leftHip"],
  mouth: ["leftMouth", "rightMouth"],
  leftLegUpper: ["leftHip", "leftKnee"],
  leftLegLower: ["leftKnee", "leftAnkle"],
  leftFoot: ["leftAnkle", "leftHeel", "leftFoot"],
  leftTorso: ["leftShoulder", "leftHip"],
  leftArmUpper: ["leftShoulder", "leftElbow"],
  leftArmLower: ["leftElbow", "leftWrist"],
  leftHand: ["leftWrist", "leftPalm"],
  leftHandPinky: ["leftPalm", "leftPinky"],
  leftHandIndex: ["leftPalm", "leftIndex"],
  leftHandThumb: ["leftPalm", "leftThumb"],
  leftEyeOutline: ["leftEyeInside", "leftEyeOutside"],
  rightLegUpper: ["rightHip", "rightKnee"],
  rightLegLower: ["rightKnee", "rightAnkle"],
  rightFoot: ["rightAnkle", "rightHeel", "rightFoot"],
  rightTorso: ["rightShoulder", "rightHip"],
  rightArmUpper: ["rightShoulder", "rightElbow"],
  rightArmLower: ["rightElbow", "rightWrist"],
  rightHand: ["rightWrist", "rightPalm"],
  rightHandPinky: ["rightPalm", "rightPinky"],
  rightHandIndex: ["rightPalm", "rightIndex"],
  rightHandThumb: ["rightPalm", "rightThumb"],
  rightEyeOutline: ["rightEyeInside", "rightEyeOutside"]
};

// src/body/blazeposedetector.ts
var inputSize2 = 224;
var anchorTensor;
var numLayers = 5;
var strides = [8, 16, 32, 32, 32];
async function createAnchors() {
  const anchors3 = [];
  let layerId = 0;
  while (layerId < numLayers) {
    let anchorCount = 0;
    let lastSameStrideLayer = layerId;
    while (lastSameStrideLayer < strides.length && strides[lastSameStrideLayer] === strides[layerId]) {
      anchorCount += 2;
      lastSameStrideLayer++;
    }
    const stride = strides[layerId];
    const featureMapHeight = Math.ceil(inputSize2 / stride);
    const featureMapWidth = Math.ceil(inputSize2 / stride);
    for (let y = 0; y < featureMapHeight; ++y) {
      for (let x = 0; x < featureMapWidth; ++x) {
        for (let anchorId = 0; anchorId < anchorCount; ++anchorId) {
          anchors3.push({ x: (x + 0.5) / featureMapWidth, y: (y + 0.5) / featureMapHeight });
        }
      }
    }
    layerId = lastSameStrideLayer;
  }
  anchorTensor = { x: Zt(anchors3.map((a) => a.x)), y: Zt(anchors3.map((a) => a.y)) };
}

// src/util/box.ts
function calc(keypoints, outputSize2 = [1, 1]) {
  const coords = [keypoints.map((pt2) => pt2[0]), keypoints.map((pt2) => pt2[1])];
  const min = [Math.min(...coords[0]), Math.min(...coords[1])];
  const max = [Math.max(...coords[0]), Math.max(...coords[1])];
  const box = [min[0], min[1], max[0] - min[0], max[1] - min[1]];
  const boxRaw = [box[0] / outputSize2[0], box[1] / outputSize2[1], box[2] / outputSize2[0], box[3] / outputSize2[1]];
  return { box, boxRaw };
}
function square(keypoints, outputSize2 = [1, 1]) {
  const coords = [keypoints.map((pt2) => pt2[0]), keypoints.map((pt2) => pt2[1])];
  const min = [Math.min(...coords[0]), Math.min(...coords[1])];
  const max = [Math.max(...coords[0]), Math.max(...coords[1])];
  const center = [(min[0] + max[0]) / 2, (min[1] + max[1]) / 2];
  const dist = Math.max(center[0] - min[0], center[1] - min[1], -center[0] + max[0], -center[1] + max[1]);
  const box = [Math.trunc(center[0] - dist), Math.trunc(center[1] - dist), Math.trunc(2 * dist), Math.trunc(2 * dist)];
  const boxRaw = [box[0] / outputSize2[0], box[1] / outputSize2[1], box[2] / outputSize2[0], box[3] / outputSize2[1]];
  return { box, boxRaw };
}
function scale(box, scaleFact) {
  const dist = [box[2] * scaleFact, box[3] * scaleFact];
  const newBox = [
    box[0] - (dist[0] - box[2]) / 2,
    box[1] - (dist[1] - box[3]) / 2,
    dist[0],
    dist[1]
  ];
  return newBox;
}

// src/body/blazepose.ts
var env2 = { initial: true };
var models = { detector: null, landmarks: null };
var inputSize3 = { detector: [224, 224], landmarks: [256, 256] };
var skipped5 = Number.MAX_SAFE_INTEGER;
var outputNodes = {
  landmarks: ["ld_3d", "activation_segmentation", "activation_heatmap", "world_3d", "output_poseflag"],
  detector: []
};
var cache = null;
var cropBox;
var padding = [[0, 0], [0, 0], [0, 0], [0, 0]];
var lastTime5 = 0;
var sigmoid = (x) => 1 - 1 / (1 + Math.exp(x));
async function loadDetect(config3) {
  if (env2.initial)
    models.detector = null;
  if (!models.detector && config3.body["detector"] && config3.body["detector"]["modelPath"] || "") {
    models.detector = await loadModel(config3.body["detector"]["modelPath"]);
    const inputs = Object.values(models.detector.modelSignature["inputs"]);
    inputSize3.detector[0] = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[1].size) : 0;
    inputSize3.detector[1] = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[2].size) : 0;
  } else if (config3.debug && models.detector)
    log("cached model:", models.detector["modelUrl"]);
  await createAnchors();
  return models.detector;
}
async function loadPose(config3) {
  if (env2.initial)
    models.landmarks = null;
  if (!models.landmarks) {
    models.landmarks = await loadModel(config3.body.modelPath);
    const inputs = Object.values(models.landmarks.modelSignature["inputs"]);
    inputSize3.landmarks[0] = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[1].size) : 0;
    inputSize3.landmarks[1] = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[2].size) : 0;
  } else if (config3.debug)
    log("cached model:", models.landmarks["modelUrl"]);
  return models.landmarks;
}
async function prepareImage(input, size2) {
  const t = {};
  if (!input.shape || !input.shape[1] || !input.shape[2])
    return input;
  let final;
  if (cropBox) {
    t.cropped = jn.cropAndResize(input, [cropBox], [0], [input.shape[1], input.shape[2]]);
  }
  if (input.shape[1] !== input.shape[2]) {
    const height = [
      input.shape[2] > input.shape[1] ? Math.trunc((input.shape[2] - input.shape[1]) / 2) : 0,
      input.shape[2] > input.shape[1] ? Math.trunc((input.shape[2] - input.shape[1]) / 2) : 0
    ];
    const width = [
      input.shape[1] > input.shape[2] ? Math.trunc((input.shape[1] - input.shape[2]) / 2) : 0,
      input.shape[1] > input.shape[2] ? Math.trunc((input.shape[1] - input.shape[2]) / 2) : 0
    ];
    padding = [
      [0, 0],
      height,
      width,
      [0, 0]
    ];
    t.pad = bi(t.cropped || input, padding);
    t.resize = jn.resizeBilinear(t.pad, [size2, size2]);
    final = xe(t.resize, constants.tf255);
  } else if (input.shape[1] !== size2) {
    t.resize = jn.resizeBilinear(t.cropped || input, [size2, size2]);
    final = xe(t.resize, constants.tf255);
  } else {
    final = xe(t.cropped || input, constants.tf255);
  }
  Object.keys(t).forEach((tensor) => De(t[tensor]));
  return final;
}
function rescaleKeypoints(keypoints, outputSize2) {
  for (const kpt4 of keypoints) {
    kpt4.position = [
      Math.trunc(kpt4.position[0] * (outputSize2[0] + padding[2][0] + padding[2][1]) / outputSize2[0] - padding[2][0]),
      Math.trunc(kpt4.position[1] * (outputSize2[1] + padding[1][0] + padding[1][1]) / outputSize2[1] - padding[1][0]),
      kpt4.position[2]
    ];
    kpt4.positionRaw = [kpt4.position[0] / outputSize2[0], kpt4.position[1] / outputSize2[1], 2 * kpt4.position[2] / (outputSize2[0] + outputSize2[1])];
  }
  if (cropBox) {
    for (const kpt4 of keypoints) {
      kpt4.positionRaw = [
        kpt4.positionRaw[0] + cropBox[1],
        kpt4.positionRaw[1] + cropBox[0],
        kpt4.positionRaw[2]
      ];
      kpt4.position = [
        Math.trunc(kpt4.positionRaw[0] * outputSize2[0]),
        Math.trunc(kpt4.positionRaw[1] * outputSize2[1]),
        kpt4.positionRaw[2]
      ];
    }
  }
  return keypoints;
}
async function fixKeypoints(keypoints) {
  const leftPalm = keypoints.find((k) => k.part === "leftPalm");
  const leftWrist = keypoints.find((k) => k.part === "leftWrist");
  const leftIndex = keypoints.find((k) => k.part === "leftIndex");
  leftPalm.position[2] = ((leftWrist.position[2] || 0) + (leftIndex.position[2] || 0)) / 2;
  const rightPalm = keypoints.find((k) => k.part === "rightPalm");
  const rightWrist = keypoints.find((k) => k.part === "rightWrist");
  const rightIndex = keypoints.find((k) => k.part === "rightIndex");
  rightPalm.position[2] = ((rightWrist.position[2] || 0) + (rightIndex.position[2] || 0)) / 2;
}
async function detectLandmarks(input, config3, outputSize2) {
  var _a2;
  const t = {};
  [t.ld, t.segmentation, t.heatmap, t.world, t.poseflag] = (_a2 = models.landmarks) == null ? void 0 : _a2.execute(input, outputNodes.landmarks);
  const poseScore = (await t.poseflag.data())[0];
  const points = await t.ld.data();
  const distances = await t.world.data();
  Object.keys(t).forEach((tensor) => De(t[tensor]));
  const keypointsRelative = [];
  const depth = 5;
  for (let i = 0; i < points.length / depth; i++) {
    const score = sigmoid(points[depth * i + 3]);
    const presence = sigmoid(points[depth * i + 4]);
    const adjScore = Math.trunc(100 * score * presence * poseScore) / 100;
    const positionRaw = [points[depth * i + 0] / inputSize3.landmarks[0], points[depth * i + 1] / inputSize3.landmarks[1], points[depth * i + 2] + 0];
    const position = [Math.trunc(outputSize2[0] * positionRaw[0]), Math.trunc(outputSize2[1] * positionRaw[1]), positionRaw[2]];
    const distance2 = [distances[depth * i + 0], distances[depth * i + 1], distances[depth * i + 2] + 0];
    keypointsRelative.push({ part: kpt[i], positionRaw, position, distance: distance2, score: adjScore });
  }
  if (poseScore < (config3.body.minConfidence || 0))
    return null;
  fixKeypoints(keypointsRelative);
  const keypoints = rescaleKeypoints(keypointsRelative, outputSize2);
  const kpts = keypoints.map((k) => k.position);
  const boxes = calc(kpts, [outputSize2[0], outputSize2[1]]);
  const annotations2 = {};
  for (const [name, indexes] of Object.entries(connected)) {
    const pt2 = [];
    for (let i = 0; i < indexes.length - 1; i++) {
      const pt0 = keypoints.find((kpt4) => kpt4.part === indexes[i]);
      const pt1 = keypoints.find((kpt4) => kpt4.part === indexes[i + 1]);
      if (pt0 && pt1)
        pt2.push([pt0.position, pt1.position]);
    }
    annotations2[name] = pt2;
  }
  const body4 = { id: 0, score: Math.trunc(100 * poseScore) / 100, box: boxes.box, boxRaw: boxes.boxRaw, keypoints, annotations: annotations2 };
  return body4;
}
async function predict5(input, config3) {
  const outputSize2 = [input.shape[2] || 0, input.shape[1] || 0];
  const skipTime = (config3.body.skipTime || 0) > now() - lastTime5;
  const skipFrame = skipped5 < (config3.body.skipFrames || 0);
  if (config3.skipAllowed && skipTime && skipFrame && cache !== null) {
    skipped5++;
  } else {
    const t = {};
    t.landmarks = await prepareImage(input, 256);
    cache = await detectLandmarks(t.landmarks, config3, outputSize2);
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    lastTime5 = now();
    skipped5 = 0;
  }
  return cache ? [cache] : [];
}

// src/object/labels.ts
var labels = [
  { class: 1, label: "person" },
  { class: 2, label: "bicycle" },
  { class: 3, label: "car" },
  { class: 4, label: "motorcycle" },
  { class: 5, label: "airplane" },
  { class: 6, label: "bus" },
  { class: 7, label: "train" },
  { class: 8, label: "truck" },
  { class: 9, label: "boat" },
  { class: 10, label: "traffic light" },
  { class: 11, label: "fire hydrant" },
  { class: 12, label: "stop sign" },
  { class: 13, label: "parking meter" },
  { class: 14, label: "bench" },
  { class: 15, label: "bird" },
  { class: 16, label: "cat" },
  { class: 17, label: "dog" },
  { class: 18, label: "horse" },
  { class: 19, label: "sheep" },
  { class: 20, label: "cow" },
  { class: 21, label: "elephant" },
  { class: 22, label: "bear" },
  { class: 23, label: "zebra" },
  { class: 24, label: "giraffe" },
  { class: 25, label: "backpack" },
  { class: 26, label: "umbrella" },
  { class: 27, label: "handbag" },
  { class: 28, label: "tie" },
  { class: 29, label: "suitcase" },
  { class: 30, label: "frisbee" },
  { class: 31, label: "skis" },
  { class: 32, label: "snowboard" },
  { class: 33, label: "sports ball" },
  { class: 34, label: "kite" },
  { class: 35, label: "baseball bat" },
  { class: 36, label: "baseball glove" },
  { class: 37, label: "skateboard" },
  { class: 38, label: "surfboard" },
  { class: 39, label: "tennis racket" },
  { class: 40, label: "bottle" },
  { class: 41, label: "wine glass" },
  { class: 42, label: "cup" },
  { class: 43, label: "fork" },
  { class: 44, label: "knife" },
  { class: 45, label: "spoon" },
  { class: 46, label: "bowl" },
  { class: 47, label: "banana" },
  { class: 48, label: "apple" },
  { class: 49, label: "sandwich" },
  { class: 50, label: "orange" },
  { class: 51, label: "broccoli" },
  { class: 52, label: "carrot" },
  { class: 53, label: "hot dog" },
  { class: 54, label: "pizza" },
  { class: 55, label: "donut" },
  { class: 56, label: "cake" },
  { class: 57, label: "chair" },
  { class: 58, label: "couch" },
  { class: 59, label: "potted plant" },
  { class: 60, label: "bed" },
  { class: 61, label: "dining table" },
  { class: 62, label: "toilet" },
  { class: 63, label: "tv" },
  { class: 64, label: "laptop" },
  { class: 65, label: "mouse" },
  { class: 66, label: "remote" },
  { class: 67, label: "keyboard" },
  { class: 68, label: "cell phone" },
  { class: 69, label: "microwave" },
  { class: 70, label: "oven" },
  { class: 71, label: "toaster" },
  { class: 72, label: "sink" },
  { class: 73, label: "refrigerator" },
  { class: 74, label: "book" },
  { class: 75, label: "clock" },
  { class: 76, label: "vase" },
  { class: 77, label: "scissors" },
  { class: 78, label: "teddy bear" },
  { class: 79, label: "hair drier" },
  { class: 80, label: "toothbrush" }
];

// src/object/centernet.ts
var model6;
var inputSize4 = 0;
var last5 = [];
var lastTime6 = 0;
var skipped6 = Number.MAX_SAFE_INTEGER;
async function load6(config3) {
  if (env.initial)
    model6 = null;
  if (!model6) {
    model6 = await loadModel(config3.object.modelPath);
    const inputs = Object.values(model6.modelSignature["inputs"]);
    inputSize4 = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[2].size) : 0;
  } else if (config3.debug)
    log("cached model:", model6["modelUrl"]);
  return model6;
}
async function process3(res, outputShape, config3) {
  if (!res)
    return [];
  const t = {};
  const results = [];
  const detections = await res.array();
  t.squeeze = mr(res);
  const arr = Bn(t.squeeze, 6, 1);
  t.stack = es([arr[1], arr[0], arr[3], arr[2]], 1);
  t.boxes = mr(t.stack);
  t.scores = mr(arr[4]);
  t.classes = mr(arr[5]);
  De([res, ...arr]);
  t.nms = await jn.nonMaxSuppressionAsync(t.boxes, t.scores, config3.object.maxDetected, config3.object.iouThreshold, config3.object.minConfidence || 0);
  const nms = await t.nms.data();
  let i = 0;
  for (const id2 of Array.from(nms)) {
    const score = Math.trunc(100 * detections[0][id2][4]) / 100;
    const classVal = detections[0][id2][5];
    const label = labels[classVal].label;
    const [x, y] = [
      detections[0][id2][0] / inputSize4,
      detections[0][id2][1] / inputSize4
    ];
    const boxRaw = [
      x,
      y,
      detections[0][id2][2] / inputSize4 - x,
      detections[0][id2][3] / inputSize4 - y
    ];
    const box = [
      Math.trunc(boxRaw[0] * outputShape[0]),
      Math.trunc(boxRaw[1] * outputShape[1]),
      Math.trunc(boxRaw[2] * outputShape[0]),
      Math.trunc(boxRaw[3] * outputShape[1])
    ];
    results.push({ id: i++, score, class: classVal, label, box, boxRaw });
  }
  Object.keys(t).forEach((tensor) => De(t[tensor]));
  return results;
}
async function predict6(input, config3) {
  const skipTime = (config3.object.skipTime || 0) > now() - lastTime6;
  const skipFrame = skipped6 < (config3.object.skipFrames || 0);
  if (config3.skipAllowed && skipTime && skipFrame && last5.length > 0) {
    skipped6++;
    return last5;
  }
  skipped6 = 0;
  return new Promise(async (resolve) => {
    const outputSize2 = [input.shape[2] || 0, input.shape[1] || 0];
    const resize = jn.resizeBilinear(input, [inputSize4, inputSize4]);
    const objectT = config3.object.enabled ? model6 == null ? void 0 : model6.execute(resize, ["tower_0/detections"]) : null;
    lastTime6 = now();
    De(resize);
    const obj = await process3(objectT, outputSize2, config3);
    last5 = obj;
    resolve(obj);
  });
}

// src/body/efficientposecoords.ts
var efficientposecoords_exports = {};
__export(efficientposecoords_exports, {
  connected: () => connected2,
  kpt: () => kpt2
});
var kpt2 = [
  "head",
  "neck",
  "rightShoulder",
  "rightElbow",
  "rightWrist",
  "chest",
  "leftShoulder",
  "leftElbow",
  "leftWrist",
  "bodyCenter",
  "rightHip",
  "rightKnee",
  "rightAnkle",
  "leftHip",
  "leftKnee",
  "leftAnkle"
];
var connected2 = {
  leftLeg: ["leftHip", "leftKnee", "leftAnkle"],
  rightLeg: ["rightHip", "rightKnee", "rightAnkle"],
  torso: ["leftShoulder", "rightShoulder", "rightHip", "leftHip", "leftShoulder"],
  leftArm: ["leftShoulder", "leftElbow", "leftWrist"],
  rightArm: ["rightShoulder", "rightElbow", "rightWrist"],
  head: []
};

// src/body/efficientpose.ts
var model7;
var lastTime7 = 0;
var cache2 = { id: 0, keypoints: [], box: [0, 0, 0, 0], boxRaw: [0, 0, 0, 0], score: 0, annotations: {} };
var skipped7 = Number.MAX_SAFE_INTEGER;
async function load7(config3) {
  if (env.initial)
    model7 = null;
  if (!model7)
    model7 = await loadModel(config3.body.modelPath);
  else if (config3.debug)
    log("cached model:", model7["modelUrl"]);
  return model7;
}
async function max2d(inputs, minScore) {
  const [width, height] = inputs.shape;
  const reshaped = U(inputs, [height * width]);
  const max = As(reshaped, 0);
  const newScore = (await max.data())[0];
  De([reshaped, max]);
  if (newScore > minScore) {
    const coordinates = Xu(reshaped, 0);
    const mod = zD(coordinates, width);
    const x = (await mod.data())[0];
    const div = xe(coordinates, we(width, "int32"));
    const y = (await div.data())[0];
    De([mod, div]);
    return [x, y, newScore];
  }
  return [0, 0, newScore];
}
async function predict7(image, config3) {
  const skipTime = (config3.body.skipTime || 0) > now() - lastTime7;
  const skipFrame = skipped7 < (config3.body.skipFrames || 0);
  if (config3.skipAllowed && skipTime && skipFrame && Object.keys(cache2.keypoints).length > 0) {
    skipped7++;
    return [cache2];
  }
  skipped7 = 0;
  return new Promise(async (resolve) => {
    var _a2;
    const tensor = j(() => {
      if (!(model7 == null ? void 0 : model7.inputs[0].shape))
        return null;
      const resize = jn.resizeBilinear(image, [model7.inputs[0].shape[2], model7.inputs[0].shape[1]], false);
      const enhance2 = V(resize, constants.tf2);
      const norm = ge(enhance2, constants.tf1);
      return norm;
    });
    let resT;
    if (config3.body.enabled)
      resT = model7 == null ? void 0 : model7.execute(tensor);
    lastTime7 = now();
    De(tensor);
    if (resT) {
      cache2.keypoints.length = 0;
      const squeeze = resT.squeeze();
      De(resT);
      const stack = squeeze.unstack(2);
      De(squeeze);
      for (let id2 = 0; id2 < stack.length; id2++) {
        const [x10, y10, partScore] = await max2d(stack[id2], config3.body.minConfidence);
        if (partScore > (((_a2 = config3.body) == null ? void 0 : _a2.minConfidence) || 0)) {
          cache2.keypoints.push({
            score: Math.round(100 * partScore) / 100,
            part: kpt2[id2],
            positionRaw: [
              x10 / model7.inputs[0].shape[2],
              y10 / model7.inputs[0].shape[1]
            ],
            position: [
              Math.round(image.shape[2] * x10 / model7.inputs[0].shape[2]),
              Math.round(image.shape[1] * y10 / model7.inputs[0].shape[1])
            ]
          });
        }
      }
      stack.forEach((s) => De(s));
    }
    cache2.score = cache2.keypoints.reduce((prev, curr) => curr.score > prev ? curr.score : prev, 0);
    const x = cache2.keypoints.map((a) => a.position[0]);
    const y = cache2.keypoints.map((a) => a.position[1]);
    cache2.box = [
      Math.min(...x),
      Math.min(...y),
      Math.max(...x) - Math.min(...x),
      Math.max(...y) - Math.min(...y)
    ];
    const xRaw = cache2.keypoints.map((a) => a.positionRaw[0]);
    const yRaw = cache2.keypoints.map((a) => a.positionRaw[1]);
    cache2.boxRaw = [
      Math.min(...xRaw),
      Math.min(...yRaw),
      Math.max(...xRaw) - Math.min(...xRaw),
      Math.max(...yRaw) - Math.min(...yRaw)
    ];
    for (const [name, indexes] of Object.entries(connected2)) {
      const pt2 = [];
      for (let i = 0; i < indexes.length - 1; i++) {
        const pt0 = cache2.keypoints.find((kpt4) => kpt4.part === indexes[i]);
        const pt1 = cache2.keypoints.find((kpt4) => kpt4.part === indexes[i + 1]);
        if (pt0 && pt1 && pt0.score > (config3.body.minConfidence || 0) && pt1.score > (config3.body.minConfidence || 0))
          pt2.push([pt0.position, pt1.position]);
      }
      cache2.annotations[name] = pt2;
    }
    resolve([cache2]);
  });
}

// src/gear/emotion.ts
var annotations = ["angry", "disgust", "fear", "happy", "sad", "surprise", "neutral"];
var model8;
var last6 = [];
var lastCount5 = 0;
var lastTime8 = 0;
var skipped8 = Number.MAX_SAFE_INTEGER;
async function load8(config3) {
  var _a2;
  if (env.initial)
    model8 = null;
  if (!model8)
    model8 = await loadModel((_a2 = config3.face.emotion) == null ? void 0 : _a2.modelPath);
  else if (config3.debug)
    log("cached model:", model8["modelUrl"]);
  return model8;
}
async function predict8(image, config3, idx, count2) {
  var _a2, _b2;
  if (!model8)
    return [];
  const skipFrame = skipped8 < (((_a2 = config3.face.emotion) == null ? void 0 : _a2.skipFrames) || 0);
  const skipTime = (((_b2 = config3.face.emotion) == null ? void 0 : _b2.skipTime) || 0) > now() - lastTime8;
  if (config3.skipAllowed && skipTime && skipFrame && lastCount5 === count2 && last6[idx] && last6[idx].length > 0) {
    skipped8++;
    return last6[idx];
  }
  skipped8 = 0;
  return new Promise(async (resolve) => {
    var _a3, _b3;
    const obj = [];
    if ((_a3 = config3.face.emotion) == null ? void 0 : _a3.enabled) {
      const t = {};
      const inputSize10 = (model8 == null ? void 0 : model8.inputs[0].shape) ? model8.inputs[0].shape[2] : 0;
      t.resize = jn.resizeBilinear(image, [inputSize10, inputSize10], false);
      t.channels = V(t.resize, constants.rgb);
      t.grayscale = ve(t.channels, 3, true);
      t.grayscaleSub = ge(t.grayscale, constants.tf05);
      t.grayscaleMul = V(t.grayscaleSub, constants.tf2);
      t.emotion = model8 == null ? void 0 : model8.execute(t.grayscaleMul);
      lastTime8 = now();
      const data = await t.emotion.data();
      for (let i = 0; i < data.length; i++) {
        if (data[i] > (((_b3 = config3.face.emotion) == null ? void 0 : _b3.minConfidence) || 0))
          obj.push({ score: Math.min(0.99, Math.trunc(100 * data[i]) / 100), emotion: annotations[i] });
      }
      obj.sort((a, b) => b.score - a.score);
      Object.keys(t).forEach((tensor) => De(t[tensor]));
    }
    last6[idx] = obj;
    lastCount5 = count2;
    resolve(obj);
  });
}

// src/face/mobilefacenet.ts
var model9;
var last7 = [];
var lastCount6 = 0;
var lastTime9 = 0;
var skipped9 = Number.MAX_SAFE_INTEGER;
async function load9(config3) {
  if (env.initial)
    model9 = null;
  if (!model9)
    model9 = await loadModel(config3.face["mobilefacenet"].modelPath);
  else if (config3.debug)
    log("cached model:", model9["modelUrl"]);
  return model9;
}
async function predict9(input, config3, idx, count2) {
  var _a2, _b2;
  if (!model9)
    return [];
  const skipFrame = skipped9 < (((_a2 = config3.face["embedding"]) == null ? void 0 : _a2.skipFrames) || 0);
  const skipTime = (((_b2 = config3.face["embedding"]) == null ? void 0 : _b2.skipTime) || 0) > now() - lastTime9;
  if (config3.skipAllowed && skipTime && skipFrame && lastCount6 === count2 && last7[idx]) {
    skipped9++;
    return last7[idx];
  }
  return new Promise(async (resolve) => {
    var _a3;
    let data = [];
    if (((_a3 = config3.face["embedding"]) == null ? void 0 : _a3.enabled) && (model9 == null ? void 0 : model9.inputs[0].shape)) {
      const t = {};
      t.crop = jn.resizeBilinear(input, [model9.inputs[0].shape[2], model9.inputs[0].shape[1]], false);
      t.data = model9 == null ? void 0 : model9.execute(t.crop);
      const output = await t.data.data();
      data = Array.from(output);
    }
    last7[idx] = data;
    lastCount6 = count2;
    lastTime9 = now();
    resolve(data);
  });
}

// src/face/iris.ts
var model10;
var inputSize5 = 0;
var irisEnlarge = 2.3;
var leftOutline = meshAnnotations["leftEyeLower0"];
var rightOutline = meshAnnotations["rightEyeLower0"];
var eyeLandmarks = {
  leftBounds: [leftOutline[0], leftOutline[leftOutline.length - 1]],
  rightBounds: [rightOutline[0], rightOutline[rightOutline.length - 1]]
};
var irisLandmarks = {
  upperCenter: 3,
  lowerCenter: 4,
  index: 71,
  numCoordinates: 76
};
async function load10(config3) {
  var _a2;
  if (env.initial)
    model10 = null;
  if (!model10)
    model10 = await loadModel((_a2 = config3.face.iris) == null ? void 0 : _a2.modelPath);
  else if (config3.debug)
    log("cached model:", model10["modelUrl"]);
  inputSize5 = model10.inputs[0].shape ? model10.inputs[0].shape[2] : 0;
  if (inputSize5 === -1)
    inputSize5 = 64;
  return model10;
}
function replaceIrisCoords(rawCoords, newCoords, prefix, keys) {
  for (let i = 0; i < irisIndices.length; i++) {
    const { key, indices } = irisIndices[i];
    const originalIndices = meshAnnotations[`${prefix}${key}`];
    if (!keys || keys.includes(key)) {
      for (let j10 = 0; j10 < indices.length; j10++) {
        const index2 = indices[j10];
        rawCoords[originalIndices[j10]] = [
          newCoords[index2][0],
          newCoords[index2][1],
          (newCoords[index2][2] + rawCoords[originalIndices[j10]][2]) / 2
        ];
      }
    }
  }
}
var getLeftToRightEyeDepthDifference = (rawCoords) => {
  const leftEyeZ = rawCoords[eyeLandmarks.leftBounds[0]][2];
  const rightEyeZ = rawCoords[eyeLandmarks.rightBounds[0]][2];
  return leftEyeZ - rightEyeZ;
};
var getEyeBox = (rawCoords, face4, eyeInnerCornerIndex, eyeOuterCornerIndex, meshSize, flip = false) => {
  const box = squarifyBox(enlargeBox(calculateLandmarksBoundingBox([rawCoords[eyeInnerCornerIndex], rawCoords[eyeOuterCornerIndex]]), irisEnlarge));
  const boxSize = getBoxSize(box);
  let crop = jn.cropAndResize(face4, [[
    box.startPoint[1] / meshSize,
    box.startPoint[0] / meshSize,
    box.endPoint[1] / meshSize,
    box.endPoint[0] / meshSize
  ]], [0], [inputSize5, inputSize5]);
  if (flip && env.kernels.includes("flipleftright")) {
    const flipped = jn.flipLeftRight(crop);
    De(crop);
    crop = flipped;
  }
  return { box, boxSize, crop };
};
var getEyeCoords = (eyeData, eyeBox, eyeBoxSize, flip = false) => {
  const eyeRawCoords = [];
  for (let i = 0; i < irisLandmarks.numCoordinates; i++) {
    const x = eyeData[i * 3];
    const y = eyeData[i * 3 + 1];
    const z = eyeData[i * 3 + 2];
    eyeRawCoords.push([
      (flip ? 1 - x / inputSize5 : x / inputSize5) * eyeBoxSize[0] + eyeBox.startPoint[0],
      y / inputSize5 * eyeBoxSize[1] + eyeBox.startPoint[1],
      z
    ]);
  }
  return { rawCoords: eyeRawCoords, iris: eyeRawCoords.slice(irisLandmarks.index) };
};
var getAdjustedIrisCoords = (rawCoords, irisCoords, direction) => {
  const upperCenterZ = rawCoords[meshAnnotations[`${direction}EyeUpper0`][irisLandmarks.upperCenter]][2];
  const lowerCenterZ = rawCoords[meshAnnotations[`${direction}EyeLower0`][irisLandmarks.lowerCenter]][2];
  const averageZ = (upperCenterZ + lowerCenterZ) / 2;
  return irisCoords.map((coord, i) => {
    let z = averageZ;
    if (i === 2) {
      z = upperCenterZ;
    } else if (i === 4) {
      z = lowerCenterZ;
    }
    return [coord[0], coord[1], z];
  });
};
async function augmentIris(rawCoords, face4, config3, meshSize) {
  if (!model10) {
    if (config3.debug)
      log("face mesh iris detection requested, but model is not loaded");
    return rawCoords;
  }
  const { box: leftEyeBox, boxSize: leftEyeBoxSize, crop: leftEyeCrop } = getEyeBox(rawCoords, face4, eyeLandmarks.leftBounds[0], eyeLandmarks.leftBounds[1], meshSize, true);
  const { box: rightEyeBox, boxSize: rightEyeBoxSize, crop: rightEyeCrop } = getEyeBox(rawCoords, face4, eyeLandmarks.rightBounds[0], eyeLandmarks.rightBounds[1], meshSize, true);
  const combined = Ot([leftEyeCrop, rightEyeCrop]);
  De(leftEyeCrop);
  De(rightEyeCrop);
  const eyePredictions = model10.execute(combined);
  De(combined);
  const eyePredictionsData = await eyePredictions.data();
  De(eyePredictions);
  const leftEyeData = eyePredictionsData.slice(0, irisLandmarks.numCoordinates * 3);
  const { rawCoords: leftEyeRawCoords, iris: leftIrisRawCoords } = getEyeCoords(leftEyeData, leftEyeBox, leftEyeBoxSize, true);
  const rightEyeData = eyePredictionsData.slice(irisLandmarks.numCoordinates * 3);
  const { rawCoords: rightEyeRawCoords, iris: rightIrisRawCoords } = getEyeCoords(rightEyeData, rightEyeBox, rightEyeBoxSize, false);
  const leftToRightEyeDepthDifference = getLeftToRightEyeDepthDifference(rawCoords);
  if (Math.abs(leftToRightEyeDepthDifference) < 30) {
    replaceIrisCoords(rawCoords, leftEyeRawCoords, "left", null);
    replaceIrisCoords(rawCoords, rightEyeRawCoords, "right", null);
  } else if (leftToRightEyeDepthDifference < 1) {
    replaceIrisCoords(rawCoords, leftEyeRawCoords, "left", ["EyeUpper0", "EyeLower0"]);
  } else {
    replaceIrisCoords(rawCoords, rightEyeRawCoords, "right", ["EyeUpper0", "EyeLower0"]);
  }
  const adjustedLeftIrisCoords = getAdjustedIrisCoords(rawCoords, leftIrisRawCoords, "left");
  const adjustedRightIrisCoords = getAdjustedIrisCoords(rawCoords, rightIrisRawCoords, "right");
  const newCoords = rawCoords.concat(adjustedLeftIrisCoords).concat(adjustedRightIrisCoords);
  return newCoords;
}

// src/face/attention.ts
var attentionDefinitions = {
  eyeLLower: [33, 7, 163, 144, 145, 153, 154, 155, 133],
  eyeRLower: [263, 249, 390, 373, 374, 380, 381, 382, 362],
  lips: [185, 96, 90, 181, 84, 17, 314, 405, 320, 307, 409, 40, 39, 73, 37, 0, 267, 269, 270, 409, 40, 88, 178, 178, 87, 14, 268, 402, 318, 324, 409, 80, 41, 38, 87, 12, 268, 303, 318, 324, 185, 95, 80, 81, 85, 16, 315, 404, 319, 325, 409, 40, 39, 73, 72, 0, 302, 303, 270, 408, 185, 88, 88, 81, 82, 15, 316, 403, 319, 324, 409, 80, 41, 38, 87, 12, 268, 303, 318, 324],
  eyeL: [33, 7, 163, 144, 145, 153, 154, 155, 133, 246, 161, 160, 159, 158, 157, 173, 130, 25, 110, 24, 23, 22, 26, 112, 243, 247, 30, 29, 27, 28, 56, 190, 226, 31, 228, 229, 230, 231, 232, 233, 244, 113, 225, 224, 223, 222, 221, 189, 35, 124, 46, 53, 52, 65, 143, 111, 117, 118, 119, 120, 121, 128, 245, 156, 70, 63, 105, 66, 107, 55, 193],
  eyeR: [263, 249, 390, 373, 374, 380, 381, 382, 362, 466, 388, 387, 386, 385, 384, 398, 359, 255, 339, 254, 253, 252, 256, 341, 463, 467, 260, 259, 257, 258, 286, 414, 446, 261, 448, 449, 450, 451, 452, 453, 464, 342, 445, 444, 443, 442, 441, 413, 265, 353, 276, 283, 282, 295, 372, 340, 346, 347, 348, 349, 350, 357, 465, 383, 300, 293, 334, 296, 336, 285, 417]
};
async function augment(rawCoords, results) {
  const t = {
    irisL: results[3].dataSync(),
    irisR: results[1].dataSync(),
    eyeL: results[0].dataSync(),
    eyeR: results[6].dataSync(),
    lips: results[5].dataSync()
  };
  const irisRDepth = attentionDefinitions.eyeRLower.reduce((prev, curr) => prev += rawCoords[curr][2], 0) / attentionDefinitions.eyeRLower.length;
  for (let i = 0; i < t.irisR.length / 2; i++)
    rawCoords.push([t.irisR[2 * i + 0], t.irisR[2 * i + 1], irisRDepth]);
  const irisLDepth = attentionDefinitions.eyeLLower.reduce((prev, curr) => prev += rawCoords[curr][2], 0) / attentionDefinitions.eyeLLower.length;
  for (let i = 0; i < t.irisL.length / 2; i++)
    rawCoords.push([t.irisL[2 * i + 0], t.irisL[2 * i + 1], irisLDepth]);
  for (let i = 0; i < t.eyeL.length / 2; i++)
    rawCoords[attentionDefinitions.eyeL[i]] = [t.eyeL[2 * i + 0], t.eyeL[2 * i + 1], rawCoords[attentionDefinitions.eyeL[i]][2]];
  for (let i = 0; i < t.eyeR.length / 2; i++)
    rawCoords[attentionDefinitions.eyeR[i]] = [t.eyeR[2 * i + 0], t.eyeR[2 * i + 1], rawCoords[attentionDefinitions.eyeR[i]][2]];
  for (let i = 0; i < t.lips.length / 2; i++)
    rawCoords[attentionDefinitions.lips[i]] = [t.lips[2 * i + 0], t.lips[2 * i + 1], rawCoords[attentionDefinitions.lips[i]][2]];
  return rawCoords;
}

// src/face/facemesh.ts
var cache3 = {
  boxes: [],
  skipped: Number.MAX_SAFE_INTEGER,
  timestamp: 0
};
var model11 = null;
var inputSize6 = 0;
async function predict10(input, config3) {
  var _a2, _b2, _c, _d2, _e2, _f, _g2, _h, _i, _j2;
  const skipTime = (((_a2 = config3.face.detector) == null ? void 0 : _a2.skipTime) || 0) > now() - cache3.timestamp;
  const skipFrame = cache3.skipped < (((_b2 = config3.face.detector) == null ? void 0 : _b2.skipFrames) || 0);
  if (!config3.skipAllowed || !skipTime || !skipFrame || cache3.boxes.length === 0) {
    cache3.boxes = await getBoxes(input, config3);
    cache3.timestamp = now();
    cache3.skipped = 0;
  } else {
    cache3.skipped++;
  }
  const faces = [];
  const newCache = [];
  let id2 = 0;
  for (let i = 0; i < cache3.boxes.length; i++) {
    const box = cache3.boxes[i];
    let angle = 0;
    let rotationMatrix;
    const face4 = {
      id: id2++,
      mesh: [],
      meshRaw: [],
      box: [0, 0, 0, 0],
      boxRaw: [0, 0, 0, 0],
      score: 0,
      boxScore: 0,
      faceScore: 0,
      annotations: {}
    };
    [angle, rotationMatrix, face4.tensor] = correctFaceRotation((_c = config3.face.detector) == null ? void 0 : _c.rotation, box, input, ((_d2 = config3.face.mesh) == null ? void 0 : _d2.enabled) ? inputSize6 : size());
    if ((_e2 = config3 == null ? void 0 : config3.filter) == null ? void 0 : _e2.equalization) {
      const equilized = await histogramEqualization(face4.tensor);
      De(face4.tensor);
      face4.tensor = equilized;
    }
    face4.boxScore = Math.round(100 * box.confidence) / 100;
    if (!((_f = config3.face.mesh) == null ? void 0 : _f.enabled)) {
      face4.box = clampBox(box, input);
      face4.boxRaw = getRawBox(box, input);
      face4.score = face4.boxScore;
      face4.mesh = box.landmarks.map((pt2) => [
        (box.startPoint[0] + box.endPoint[0]) / 2 + (box.endPoint[0] + box.startPoint[0]) * pt2[0] / size(),
        (box.startPoint[1] + box.endPoint[1]) / 2 + (box.endPoint[1] + box.startPoint[1]) * pt2[1] / size()
      ]);
      face4.meshRaw = face4.mesh.map((pt2) => [pt2[0] / (input.shape[2] || 0), pt2[1] / (input.shape[1] || 0), (pt2[2] || 0) / inputSize6]);
      for (const key of Object.keys(blazeFaceLandmarks)) {
        face4.annotations[key] = [face4.mesh[blazeFaceLandmarks[key]]];
      }
    } else if (!model11) {
      if (config3.debug)
        log("face mesh detection requested, but model is not loaded");
    } else {
      const results = model11.execute(face4.tensor);
      const confidence = results.find((t) => t.shape[t.shape.length - 1] === 1);
      const contourCoords = results.find((t) => t.shape[t.shape.length - 1] === 1404);
      const faceConfidence = await confidence.data();
      face4.faceScore = Math.round(100 * faceConfidence[0]) / 100;
      const coordsReshaped = U(contourCoords, [-1, 3]);
      let rawCoords = await coordsReshaped.array();
      if (face4.faceScore < (((_g2 = config3.face.detector) == null ? void 0 : _g2.minConfidence) || 1)) {
        box.confidence = face4.faceScore;
      } else {
        if ((_h = config3.face.attention) == null ? void 0 : _h.enabled) {
          rawCoords = await augment(rawCoords, results);
        } else if ((_i = config3.face.iris) == null ? void 0 : _i.enabled) {
          rawCoords = await augmentIris(rawCoords, face4.tensor, config3, inputSize6);
        }
        face4.mesh = transformRawCoords(rawCoords, box, angle, rotationMatrix, inputSize6);
        face4.meshRaw = face4.mesh.map((pt2) => [pt2[0] / (input.shape[2] || 0), pt2[1] / (input.shape[1] || 0), (pt2[2] || 0) / inputSize6]);
        for (const key of Object.keys(meshAnnotations))
          face4.annotations[key] = meshAnnotations[key].map((index2) => face4.mesh[index2]);
        face4.score = face4.faceScore;
        const calculatedBox = { ...calculateFaceBox(face4.mesh, box), confidence: box.confidence, landmarks: box.landmarks };
        face4.box = clampBox(calculatedBox, input);
        face4.boxRaw = getRawBox(calculatedBox, input);
        newCache.push(calculatedBox);
      }
      De([...results, coordsReshaped]);
    }
    if (face4.score > (((_j2 = config3.face.detector) == null ? void 0 : _j2.minConfidence) || 1))
      faces.push(face4);
    else
      De(face4.tensor);
  }
  cache3.boxes = newCache;
  return faces;
}
async function load11(config3) {
  var _a2, _b2, _c;
  if (env.initial)
    model11 = null;
  if (!model11) {
    if ((_a2 = config3.face.attention) == null ? void 0 : _a2.enabled)
      model11 = await loadModel((_b2 = config3.face.attention) == null ? void 0 : _b2.modelPath);
    else
      model11 = await loadModel((_c = config3.face.mesh) == null ? void 0 : _c.modelPath);
  } else if (config3.debug) {
    log("cached model:", model11["modelUrl"]);
  }
  inputSize6 = model11.inputs[0].shape ? model11.inputs[0].shape[2] : 0;
  return model11;
}
var triangulation = TRI468;
var uvmap = UV468;

// src/face/faceres.ts
var model12;
var last8 = [];
var lastTime10 = 0;
var lastCount7 = 0;
var skipped10 = Number.MAX_SAFE_INTEGER;
async function load12(config3) {
  var _a2;
  if (env.initial)
    model12 = null;
  if (!model12)
    model12 = await loadModel((_a2 = config3.face.description) == null ? void 0 : _a2.modelPath);
  else if (config3.debug)
    log("cached model:", model12["modelUrl"]);
  return model12;
}
function enhance(input) {
  const tensor = input.image || input.tensor || input;
  if (!(model12 == null ? void 0 : model12.inputs[0].shape))
    return tensor;
  const crop = jn.resizeBilinear(tensor, [model12.inputs[0].shape[2], model12.inputs[0].shape[1]], false);
  const norm = V(crop, constants.tf255);
  De(crop);
  return norm;
}
async function predict11(image, config3, idx, count2) {
  var _a2, _b2, _c, _d2;
  if (!model12)
    return { age: 0, gender: "unknown", genderScore: 0, descriptor: [] };
  const skipFrame = skipped10 < (((_a2 = config3.face.description) == null ? void 0 : _a2.skipFrames) || 0);
  const skipTime = (((_b2 = config3.face.description) == null ? void 0 : _b2.skipTime) || 0) > now() - lastTime10;
  if (config3.skipAllowed && skipFrame && skipTime && lastCount7 === count2 && ((_c = last8[idx]) == null ? void 0 : _c.age) && ((_d2 = last8[idx]) == null ? void 0 : _d2.age) > 0) {
    skipped10++;
    return last8[idx];
  }
  skipped10 = 0;
  return new Promise(async (resolve) => {
    var _a3, _b3;
    const obj = {
      age: 0,
      gender: "unknown",
      genderScore: 0,
      descriptor: []
    };
    if ((_a3 = config3.face.description) == null ? void 0 : _a3.enabled) {
      const enhanced = enhance(image);
      const resT = model12 == null ? void 0 : model12.execute(enhanced);
      lastTime10 = now();
      De(enhanced);
      const genderT = await resT.find((t) => t.shape[1] === 1);
      const gender = await genderT.data();
      const confidence = Math.trunc(200 * Math.abs(gender[0] - 0.5)) / 100;
      if (confidence > (((_b3 = config3.face.description) == null ? void 0 : _b3.minConfidence) || 0)) {
        obj.gender = gender[0] <= 0.5 ? "female" : "male";
        obj.genderScore = Math.min(0.99, confidence);
      }
      const argmax = Xu(resT.find((t) => t.shape[1] === 100), 1);
      const age = (await argmax.data())[0];
      De(argmax);
      const ageT = resT.find((t) => t.shape[1] === 100);
      const all2 = await ageT.data();
      obj.age = Math.round(all2[age - 1] > all2[age + 1] ? 10 * age - 100 * all2[age - 1] : 10 * age + 100 * all2[age + 1]) / 10;
      const desc = resT.find((t) => t.shape[1] === 1024);
      const descriptor = desc ? await desc.data() : [];
      obj.descriptor = Array.from(descriptor);
      resT.forEach((t) => De(t));
    }
    last8[idx] = obj;
    lastCount7 = count2;
    resolve(obj);
  });
}

// src/hand/handposeutil.ts
function getBoxSize2(box) {
  return [
    Math.abs(box.endPoint[0] - box.startPoint[0]),
    Math.abs(box.endPoint[1] - box.startPoint[1])
  ];
}
function getBoxCenter2(box) {
  return [
    box.startPoint[0] + (box.endPoint[0] - box.startPoint[0]) / 2,
    box.startPoint[1] + (box.endPoint[1] - box.startPoint[1]) / 2
  ];
}
function cutBoxFromImageAndResize(box, image, cropSize) {
  const h = image.shape[1];
  const w10 = image.shape[2];
  const boxes = [[
    box.startPoint[1] / h,
    box.startPoint[0] / w10,
    box.endPoint[1] / h,
    box.endPoint[0] / w10
  ]];
  return jn.cropAndResize(image, boxes, [0], cropSize);
}
function scaleBoxCoordinates2(box, factor) {
  const startPoint = [box.startPoint[0] * factor[0], box.startPoint[1] * factor[1]];
  const endPoint = [box.endPoint[0] * factor[0], box.endPoint[1] * factor[1]];
  const palmLandmarks = box.palmLandmarks.map((coord) => {
    const scaledCoord = [coord[0] * factor[0], coord[1] * factor[1]];
    return scaledCoord;
  });
  return { startPoint, endPoint, palmLandmarks, confidence: box.confidence };
}
function enlargeBox2(box, factor = 1.5) {
  const center = getBoxCenter2(box);
  const size2 = getBoxSize2(box);
  const newHalfSize = [factor * size2[0] / 2, factor * size2[1] / 2];
  const startPoint = [center[0] - newHalfSize[0], center[1] - newHalfSize[1]];
  const endPoint = [center[0] + newHalfSize[0], center[1] + newHalfSize[1]];
  return { startPoint, endPoint, palmLandmarks: box.palmLandmarks };
}
function squarifyBox2(box) {
  const centers = getBoxCenter2(box);
  const size2 = getBoxSize2(box);
  const maxEdge = Math.max(...size2);
  const halfSize = maxEdge / 2;
  const startPoint = [centers[0] - halfSize, centers[1] - halfSize];
  const endPoint = [centers[0] + halfSize, centers[1] + halfSize];
  return { startPoint, endPoint, palmLandmarks: box.palmLandmarks };
}
function normalizeRadians2(angle) {
  return angle - 2 * Math.PI * Math.floor((angle + Math.PI) / (2 * Math.PI));
}
function computeRotation2(point1, point2) {
  const radians = Math.PI / 2 - Math.atan2(-(point2[1] - point1[1]), point2[0] - point1[0]);
  return normalizeRadians2(radians);
}
var buildTranslationMatrix2 = (x, y) => [[1, 0, x], [0, 1, y], [0, 0, 1]];
function dot2(v12, v22) {
  let product = 0;
  for (let i = 0; i < v12.length; i++) {
    product += v12[i] * v22[i];
  }
  return product;
}
function getColumnFrom2DArr2(arr, columnIndex) {
  const column = [];
  for (let i = 0; i < arr.length; i++) {
    column.push(arr[i][columnIndex]);
  }
  return column;
}
function multiplyTransformMatrices2(mat1, mat2) {
  const product = [];
  const size2 = mat1.length;
  for (let row = 0; row < size2; row++) {
    product.push([]);
    for (let col = 0; col < size2; col++) {
      product[row].push(dot2(mat1[row], getColumnFrom2DArr2(mat2, col)));
    }
  }
  return product;
}
function buildRotationMatrix2(rotation, center) {
  const cosA = Math.cos(rotation);
  const sinA = Math.sin(rotation);
  const rotationMatrix = [[cosA, -sinA, 0], [sinA, cosA, 0], [0, 0, 1]];
  const translationMatrix = buildTranslationMatrix2(center[0], center[1]);
  const translationTimesRotation = multiplyTransformMatrices2(translationMatrix, rotationMatrix);
  const negativeTranslationMatrix = buildTranslationMatrix2(-center[0], -center[1]);
  return multiplyTransformMatrices2(translationTimesRotation, negativeTranslationMatrix);
}
function invertTransformMatrix2(matrix) {
  const rotationComponent = [[matrix[0][0], matrix[1][0]], [matrix[0][1], matrix[1][1]]];
  const translationComponent = [matrix[0][2], matrix[1][2]];
  const invertedTranslation = [
    -dot2(rotationComponent[0], translationComponent),
    -dot2(rotationComponent[1], translationComponent)
  ];
  return [
    rotationComponent[0].concat(invertedTranslation[0]),
    rotationComponent[1].concat(invertedTranslation[1]),
    [0, 0, 1]
  ];
}
function rotatePoint2(homogeneousCoordinate, rotationMatrix) {
  return [
    dot2(homogeneousCoordinate, rotationMatrix[0]),
    dot2(homogeneousCoordinate, rotationMatrix[1])
  ];
}

// src/hand/handposeanchors.ts
var anchors2 = [
  { x: 0.015625, y: 0.015625 },
  { x: 0.015625, y: 0.015625 },
  { x: 0.046875, y: 0.015625 },
  { x: 0.046875, y: 0.015625 },
  { x: 0.078125, y: 0.015625 },
  { x: 0.078125, y: 0.015625 },
  { x: 0.109375, y: 0.015625 },
  { x: 0.109375, y: 0.015625 },
  { x: 0.140625, y: 0.015625 },
  { x: 0.140625, y: 0.015625 },
  { x: 0.171875, y: 0.015625 },
  { x: 0.171875, y: 0.015625 },
  { x: 0.203125, y: 0.015625 },
  { x: 0.203125, y: 0.015625 },
  { x: 0.234375, y: 0.015625 },
  { x: 0.234375, y: 0.015625 },
  { x: 0.265625, y: 0.015625 },
  { x: 0.265625, y: 0.015625 },
  { x: 0.296875, y: 0.015625 },
  { x: 0.296875, y: 0.015625 },
  { x: 0.328125, y: 0.015625 },
  { x: 0.328125, y: 0.015625 },
  { x: 0.359375, y: 0.015625 },
  { x: 0.359375, y: 0.015625 },
  { x: 0.390625, y: 0.015625 },
  { x: 0.390625, y: 0.015625 },
  { x: 0.421875, y: 0.015625 },
  { x: 0.421875, y: 0.015625 },
  { x: 0.453125, y: 0.015625 },
  { x: 0.453125, y: 0.015625 },
  { x: 0.484375, y: 0.015625 },
  { x: 0.484375, y: 0.015625 },
  { x: 0.515625, y: 0.015625 },
  { x: 0.515625, y: 0.015625 },
  { x: 0.546875, y: 0.015625 },
  { x: 0.546875, y: 0.015625 },
  { x: 0.578125, y: 0.015625 },
  { x: 0.578125, y: 0.015625 },
  { x: 0.609375, y: 0.015625 },
  { x: 0.609375, y: 0.015625 },
  { x: 0.640625, y: 0.015625 },
  { x: 0.640625, y: 0.015625 },
  { x: 0.671875, y: 0.015625 },
  { x: 0.671875, y: 0.015625 },
  { x: 0.703125, y: 0.015625 },
  { x: 0.703125, y: 0.015625 },
  { x: 0.734375, y: 0.015625 },
  { x: 0.734375, y: 0.015625 },
  { x: 0.765625, y: 0.015625 },
  { x: 0.765625, y: 0.015625 },
  { x: 0.796875, y: 0.015625 },
  { x: 0.796875, y: 0.015625 },
  { x: 0.828125, y: 0.015625 },
  { x: 0.828125, y: 0.015625 },
  { x: 0.859375, y: 0.015625 },
  { x: 0.859375, y: 0.015625 },
  { x: 0.890625, y: 0.015625 },
  { x: 0.890625, y: 0.015625 },
  { x: 0.921875, y: 0.015625 },
  { x: 0.921875, y: 0.015625 },
  { x: 0.953125, y: 0.015625 },
  { x: 0.953125, y: 0.015625 },
  { x: 0.984375, y: 0.015625 },
  { x: 0.984375, y: 0.015625 },
  { x: 0.015625, y: 0.046875 },
  { x: 0.015625, y: 0.046875 },
  { x: 0.046875, y: 0.046875 },
  { x: 0.046875, y: 0.046875 },
  { x: 0.078125, y: 0.046875 },
  { x: 0.078125, y: 0.046875 },
  { x: 0.109375, y: 0.046875 },
  { x: 0.109375, y: 0.046875 },
  { x: 0.140625, y: 0.046875 },
  { x: 0.140625, y: 0.046875 },
  { x: 0.171875, y: 0.046875 },
  { x: 0.171875, y: 0.046875 },
  { x: 0.203125, y: 0.046875 },
  { x: 0.203125, y: 0.046875 },
  { x: 0.234375, y: 0.046875 },
  { x: 0.234375, y: 0.046875 },
  { x: 0.265625, y: 0.046875 },
  { x: 0.265625, y: 0.046875 },
  { x: 0.296875, y: 0.046875 },
  { x: 0.296875, y: 0.046875 },
  { x: 0.328125, y: 0.046875 },
  { x: 0.328125, y: 0.046875 },
  { x: 0.359375, y: 0.046875 },
  { x: 0.359375, y: 0.046875 },
  { x: 0.390625, y: 0.046875 },
  { x: 0.390625, y: 0.046875 },
  { x: 0.421875, y: 0.046875 },
  { x: 0.421875, y: 0.046875 },
  { x: 0.453125, y: 0.046875 },
  { x: 0.453125, y: 0.046875 },
  { x: 0.484375, y: 0.046875 },
  { x: 0.484375, y: 0.046875 },
  { x: 0.515625, y: 0.046875 },
  { x: 0.515625, y: 0.046875 },
  { x: 0.546875, y: 0.046875 },
  { x: 0.546875, y: 0.046875 },
  { x: 0.578125, y: 0.046875 },
  { x: 0.578125, y: 0.046875 },
  { x: 0.609375, y: 0.046875 },
  { x: 0.609375, y: 0.046875 },
  { x: 0.640625, y: 0.046875 },
  { x: 0.640625, y: 0.046875 },
  { x: 0.671875, y: 0.046875 },
  { x: 0.671875, y: 0.046875 },
  { x: 0.703125, y: 0.046875 },
  { x: 0.703125, y: 0.046875 },
  { x: 0.734375, y: 0.046875 },
  { x: 0.734375, y: 0.046875 },
  { x: 0.765625, y: 0.046875 },
  { x: 0.765625, y: 0.046875 },
  { x: 0.796875, y: 0.046875 },
  { x: 0.796875, y: 0.046875 },
  { x: 0.828125, y: 0.046875 },
  { x: 0.828125, y: 0.046875 },
  { x: 0.859375, y: 0.046875 },
  { x: 0.859375, y: 0.046875 },
  { x: 0.890625, y: 0.046875 },
  { x: 0.890625, y: 0.046875 },
  { x: 0.921875, y: 0.046875 },
  { x: 0.921875, y: 0.046875 },
  { x: 0.953125, y: 0.046875 },
  { x: 0.953125, y: 0.046875 },
  { x: 0.984375, y: 0.046875 },
  { x: 0.984375, y: 0.046875 },
  { x: 0.015625, y: 0.078125 },
  { x: 0.015625, y: 0.078125 },
  { x: 0.046875, y: 0.078125 },
  { x: 0.046875, y: 0.078125 },
  { x: 0.078125, y: 0.078125 },
  { x: 0.078125, y: 0.078125 },
  { x: 0.109375, y: 0.078125 },
  { x: 0.109375, y: 0.078125 },
  { x: 0.140625, y: 0.078125 },
  { x: 0.140625, y: 0.078125 },
  { x: 0.171875, y: 0.078125 },
  { x: 0.171875, y: 0.078125 },
  { x: 0.203125, y: 0.078125 },
  { x: 0.203125, y: 0.078125 },
  { x: 0.234375, y: 0.078125 },
  { x: 0.234375, y: 0.078125 },
  { x: 0.265625, y: 0.078125 },
  { x: 0.265625, y: 0.078125 },
  { x: 0.296875, y: 0.078125 },
  { x: 0.296875, y: 0.078125 },
  { x: 0.328125, y: 0.078125 },
  { x: 0.328125, y: 0.078125 },
  { x: 0.359375, y: 0.078125 },
  { x: 0.359375, y: 0.078125 },
  { x: 0.390625, y: 0.078125 },
  { x: 0.390625, y: 0.078125 },
  { x: 0.421875, y: 0.078125 },
  { x: 0.421875, y: 0.078125 },
  { x: 0.453125, y: 0.078125 },
  { x: 0.453125, y: 0.078125 },
  { x: 0.484375, y: 0.078125 },
  { x: 0.484375, y: 0.078125 },
  { x: 0.515625, y: 0.078125 },
  { x: 0.515625, y: 0.078125 },
  { x: 0.546875, y: 0.078125 },
  { x: 0.546875, y: 0.078125 },
  { x: 0.578125, y: 0.078125 },
  { x: 0.578125, y: 0.078125 },
  { x: 0.609375, y: 0.078125 },
  { x: 0.609375, y: 0.078125 },
  { x: 0.640625, y: 0.078125 },
  { x: 0.640625, y: 0.078125 },
  { x: 0.671875, y: 0.078125 },
  { x: 0.671875, y: 0.078125 },
  { x: 0.703125, y: 0.078125 },
  { x: 0.703125, y: 0.078125 },
  { x: 0.734375, y: 0.078125 },
  { x: 0.734375, y: 0.078125 },
  { x: 0.765625, y: 0.078125 },
  { x: 0.765625, y: 0.078125 },
  { x: 0.796875, y: 0.078125 },
  { x: 0.796875, y: 0.078125 },
  { x: 0.828125, y: 0.078125 },
  { x: 0.828125, y: 0.078125 },
  { x: 0.859375, y: 0.078125 },
  { x: 0.859375, y: 0.078125 },
  { x: 0.890625, y: 0.078125 },
  { x: 0.890625, y: 0.078125 },
  { x: 0.921875, y: 0.078125 },
  { x: 0.921875, y: 0.078125 },
  { x: 0.953125, y: 0.078125 },
  { x: 0.953125, y: 0.078125 },
  { x: 0.984375, y: 0.078125 },
  { x: 0.984375, y: 0.078125 },
  { x: 0.015625, y: 0.109375 },
  { x: 0.015625, y: 0.109375 },
  { x: 0.046875, y: 0.109375 },
  { x: 0.046875, y: 0.109375 },
  { x: 0.078125, y: 0.109375 },
  { x: 0.078125, y: 0.109375 },
  { x: 0.109375, y: 0.109375 },
  { x: 0.109375, y: 0.109375 },
  { x: 0.140625, y: 0.109375 },
  { x: 0.140625, y: 0.109375 },
  { x: 0.171875, y: 0.109375 },
  { x: 0.171875, y: 0.109375 },
  { x: 0.203125, y: 0.109375 },
  { x: 0.203125, y: 0.109375 },
  { x: 0.234375, y: 0.109375 },
  { x: 0.234375, y: 0.109375 },
  { x: 0.265625, y: 0.109375 },
  { x: 0.265625, y: 0.109375 },
  { x: 0.296875, y: 0.109375 },
  { x: 0.296875, y: 0.109375 },
  { x: 0.328125, y: 0.109375 },
  { x: 0.328125, y: 0.109375 },
  { x: 0.359375, y: 0.109375 },
  { x: 0.359375, y: 0.109375 },
  { x: 0.390625, y: 0.109375 },
  { x: 0.390625, y: 0.109375 },
  { x: 0.421875, y: 0.109375 },
  { x: 0.421875, y: 0.109375 },
  { x: 0.453125, y: 0.109375 },
  { x: 0.453125, y: 0.109375 },
  { x: 0.484375, y: 0.109375 },
  { x: 0.484375, y: 0.109375 },
  { x: 0.515625, y: 0.109375 },
  { x: 0.515625, y: 0.109375 },
  { x: 0.546875, y: 0.109375 },
  { x: 0.546875, y: 0.109375 },
  { x: 0.578125, y: 0.109375 },
  { x: 0.578125, y: 0.109375 },
  { x: 0.609375, y: 0.109375 },
  { x: 0.609375, y: 0.109375 },
  { x: 0.640625, y: 0.109375 },
  { x: 0.640625, y: 0.109375 },
  { x: 0.671875, y: 0.109375 },
  { x: 0.671875, y: 0.109375 },
  { x: 0.703125, y: 0.109375 },
  { x: 0.703125, y: 0.109375 },
  { x: 0.734375, y: 0.109375 },
  { x: 0.734375, y: 0.109375 },
  { x: 0.765625, y: 0.109375 },
  { x: 0.765625, y: 0.109375 },
  { x: 0.796875, y: 0.109375 },
  { x: 0.796875, y: 0.109375 },
  { x: 0.828125, y: 0.109375 },
  { x: 0.828125, y: 0.109375 },
  { x: 0.859375, y: 0.109375 },
  { x: 0.859375, y: 0.109375 },
  { x: 0.890625, y: 0.109375 },
  { x: 0.890625, y: 0.109375 },
  { x: 0.921875, y: 0.109375 },
  { x: 0.921875, y: 0.109375 },
  { x: 0.953125, y: 0.109375 },
  { x: 0.953125, y: 0.109375 },
  { x: 0.984375, y: 0.109375 },
  { x: 0.984375, y: 0.109375 },
  { x: 0.015625, y: 0.140625 },
  { x: 0.015625, y: 0.140625 },
  { x: 0.046875, y: 0.140625 },
  { x: 0.046875, y: 0.140625 },
  { x: 0.078125, y: 0.140625 },
  { x: 0.078125, y: 0.140625 },
  { x: 0.109375, y: 0.140625 },
  { x: 0.109375, y: 0.140625 },
  { x: 0.140625, y: 0.140625 },
  { x: 0.140625, y: 0.140625 },
  { x: 0.171875, y: 0.140625 },
  { x: 0.171875, y: 0.140625 },
  { x: 0.203125, y: 0.140625 },
  { x: 0.203125, y: 0.140625 },
  { x: 0.234375, y: 0.140625 },
  { x: 0.234375, y: 0.140625 },
  { x: 0.265625, y: 0.140625 },
  { x: 0.265625, y: 0.140625 },
  { x: 0.296875, y: 0.140625 },
  { x: 0.296875, y: 0.140625 },
  { x: 0.328125, y: 0.140625 },
  { x: 0.328125, y: 0.140625 },
  { x: 0.359375, y: 0.140625 },
  { x: 0.359375, y: 0.140625 },
  { x: 0.390625, y: 0.140625 },
  { x: 0.390625, y: 0.140625 },
  { x: 0.421875, y: 0.140625 },
  { x: 0.421875, y: 0.140625 },
  { x: 0.453125, y: 0.140625 },
  { x: 0.453125, y: 0.140625 },
  { x: 0.484375, y: 0.140625 },
  { x: 0.484375, y: 0.140625 },
  { x: 0.515625, y: 0.140625 },
  { x: 0.515625, y: 0.140625 },
  { x: 0.546875, y: 0.140625 },
  { x: 0.546875, y: 0.140625 },
  { x: 0.578125, y: 0.140625 },
  { x: 0.578125, y: 0.140625 },
  { x: 0.609375, y: 0.140625 },
  { x: 0.609375, y: 0.140625 },
  { x: 0.640625, y: 0.140625 },
  { x: 0.640625, y: 0.140625 },
  { x: 0.671875, y: 0.140625 },
  { x: 0.671875, y: 0.140625 },
  { x: 0.703125, y: 0.140625 },
  { x: 0.703125, y: 0.140625 },
  { x: 0.734375, y: 0.140625 },
  { x: 0.734375, y: 0.140625 },
  { x: 0.765625, y: 0.140625 },
  { x: 0.765625, y: 0.140625 },
  { x: 0.796875, y: 0.140625 },
  { x: 0.796875, y: 0.140625 },
  { x: 0.828125, y: 0.140625 },
  { x: 0.828125, y: 0.140625 },
  { x: 0.859375, y: 0.140625 },
  { x: 0.859375, y: 0.140625 },
  { x: 0.890625, y: 0.140625 },
  { x: 0.890625, y: 0.140625 },
  { x: 0.921875, y: 0.140625 },
  { x: 0.921875, y: 0.140625 },
  { x: 0.953125, y: 0.140625 },
  { x: 0.953125, y: 0.140625 },
  { x: 0.984375, y: 0.140625 },
  { x: 0.984375, y: 0.140625 },
  { x: 0.015625, y: 0.171875 },
  { x: 0.015625, y: 0.171875 },
  { x: 0.046875, y: 0.171875 },
  { x: 0.046875, y: 0.171875 },
  { x: 0.078125, y: 0.171875 },
  { x: 0.078125, y: 0.171875 },
  { x: 0.109375, y: 0.171875 },
  { x: 0.109375, y: 0.171875 },
  { x: 0.140625, y: 0.171875 },
  { x: 0.140625, y: 0.171875 },
  { x: 0.171875, y: 0.171875 },
  { x: 0.171875, y: 0.171875 },
  { x: 0.203125, y: 0.171875 },
  { x: 0.203125, y: 0.171875 },
  { x: 0.234375, y: 0.171875 },
  { x: 0.234375, y: 0.171875 },
  { x: 0.265625, y: 0.171875 },
  { x: 0.265625, y: 0.171875 },
  { x: 0.296875, y: 0.171875 },
  { x: 0.296875, y: 0.171875 },
  { x: 0.328125, y: 0.171875 },
  { x: 0.328125, y: 0.171875 },
  { x: 0.359375, y: 0.171875 },
  { x: 0.359375, y: 0.171875 },
  { x: 0.390625, y: 0.171875 },
  { x: 0.390625, y: 0.171875 },
  { x: 0.421875, y: 0.171875 },
  { x: 0.421875, y: 0.171875 },
  { x: 0.453125, y: 0.171875 },
  { x: 0.453125, y: 0.171875 },
  { x: 0.484375, y: 0.171875 },
  { x: 0.484375, y: 0.171875 },
  { x: 0.515625, y: 0.171875 },
  { x: 0.515625, y: 0.171875 },
  { x: 0.546875, y: 0.171875 },
  { x: 0.546875, y: 0.171875 },
  { x: 0.578125, y: 0.171875 },
  { x: 0.578125, y: 0.171875 },
  { x: 0.609375, y: 0.171875 },
  { x: 0.609375, y: 0.171875 },
  { x: 0.640625, y: 0.171875 },
  { x: 0.640625, y: 0.171875 },
  { x: 0.671875, y: 0.171875 },
  { x: 0.671875, y: 0.171875 },
  { x: 0.703125, y: 0.171875 },
  { x: 0.703125, y: 0.171875 },
  { x: 0.734375, y: 0.171875 },
  { x: 0.734375, y: 0.171875 },
  { x: 0.765625, y: 0.171875 },
  { x: 0.765625, y: 0.171875 },
  { x: 0.796875, y: 0.171875 },
  { x: 0.796875, y: 0.171875 },
  { x: 0.828125, y: 0.171875 },
  { x: 0.828125, y: 0.171875 },
  { x: 0.859375, y: 0.171875 },
  { x: 0.859375, y: 0.171875 },
  { x: 0.890625, y: 0.171875 },
  { x: 0.890625, y: 0.171875 },
  { x: 0.921875, y: 0.171875 },
  { x: 0.921875, y: 0.171875 },
  { x: 0.953125, y: 0.171875 },
  { x: 0.953125, y: 0.171875 },
  { x: 0.984375, y: 0.171875 },
  { x: 0.984375, y: 0.171875 },
  { x: 0.015625, y: 0.203125 },
  { x: 0.015625, y: 0.203125 },
  { x: 0.046875, y: 0.203125 },
  { x: 0.046875, y: 0.203125 },
  { x: 0.078125, y: 0.203125 },
  { x: 0.078125, y: 0.203125 },
  { x: 0.109375, y: 0.203125 },
  { x: 0.109375, y: 0.203125 },
  { x: 0.140625, y: 0.203125 },
  { x: 0.140625, y: 0.203125 },
  { x: 0.171875, y: 0.203125 },
  { x: 0.171875, y: 0.203125 },
  { x: 0.203125, y: 0.203125 },
  { x: 0.203125, y: 0.203125 },
  { x: 0.234375, y: 0.203125 },
  { x: 0.234375, y: 0.203125 },
  { x: 0.265625, y: 0.203125 },
  { x: 0.265625, y: 0.203125 },
  { x: 0.296875, y: 0.203125 },
  { x: 0.296875, y: 0.203125 },
  { x: 0.328125, y: 0.203125 },
  { x: 0.328125, y: 0.203125 },
  { x: 0.359375, y: 0.203125 },
  { x: 0.359375, y: 0.203125 },
  { x: 0.390625, y: 0.203125 },
  { x: 0.390625, y: 0.203125 },
  { x: 0.421875, y: 0.203125 },
  { x: 0.421875, y: 0.203125 },
  { x: 0.453125, y: 0.203125 },
  { x: 0.453125, y: 0.203125 },
  { x: 0.484375, y: 0.203125 },
  { x: 0.484375, y: 0.203125 },
  { x: 0.515625, y: 0.203125 },
  { x: 0.515625, y: 0.203125 },
  { x: 0.546875, y: 0.203125 },
  { x: 0.546875, y: 0.203125 },
  { x: 0.578125, y: 0.203125 },
  { x: 0.578125, y: 0.203125 },
  { x: 0.609375, y: 0.203125 },
  { x: 0.609375, y: 0.203125 },
  { x: 0.640625, y: 0.203125 },
  { x: 0.640625, y: 0.203125 },
  { x: 0.671875, y: 0.203125 },
  { x: 0.671875, y: 0.203125 },
  { x: 0.703125, y: 0.203125 },
  { x: 0.703125, y: 0.203125 },
  { x: 0.734375, y: 0.203125 },
  { x: 0.734375, y: 0.203125 },
  { x: 0.765625, y: 0.203125 },
  { x: 0.765625, y: 0.203125 },
  { x: 0.796875, y: 0.203125 },
  { x: 0.796875, y: 0.203125 },
  { x: 0.828125, y: 0.203125 },
  { x: 0.828125, y: 0.203125 },
  { x: 0.859375, y: 0.203125 },
  { x: 0.859375, y: 0.203125 },
  { x: 0.890625, y: 0.203125 },
  { x: 0.890625, y: 0.203125 },
  { x: 0.921875, y: 0.203125 },
  { x: 0.921875, y: 0.203125 },
  { x: 0.953125, y: 0.203125 },
  { x: 0.953125, y: 0.203125 },
  { x: 0.984375, y: 0.203125 },
  { x: 0.984375, y: 0.203125 },
  { x: 0.015625, y: 0.234375 },
  { x: 0.015625, y: 0.234375 },
  { x: 0.046875, y: 0.234375 },
  { x: 0.046875, y: 0.234375 },
  { x: 0.078125, y: 0.234375 },
  { x: 0.078125, y: 0.234375 },
  { x: 0.109375, y: 0.234375 },
  { x: 0.109375, y: 0.234375 },
  { x: 0.140625, y: 0.234375 },
  { x: 0.140625, y: 0.234375 },
  { x: 0.171875, y: 0.234375 },
  { x: 0.171875, y: 0.234375 },
  { x: 0.203125, y: 0.234375 },
  { x: 0.203125, y: 0.234375 },
  { x: 0.234375, y: 0.234375 },
  { x: 0.234375, y: 0.234375 },
  { x: 0.265625, y: 0.234375 },
  { x: 0.265625, y: 0.234375 },
  { x: 0.296875, y: 0.234375 },
  { x: 0.296875, y: 0.234375 },
  { x: 0.328125, y: 0.234375 },
  { x: 0.328125, y: 0.234375 },
  { x: 0.359375, y: 0.234375 },
  { x: 0.359375, y: 0.234375 },
  { x: 0.390625, y: 0.234375 },
  { x: 0.390625, y: 0.234375 },
  { x: 0.421875, y: 0.234375 },
  { x: 0.421875, y: 0.234375 },
  { x: 0.453125, y: 0.234375 },
  { x: 0.453125, y: 0.234375 },
  { x: 0.484375, y: 0.234375 },
  { x: 0.484375, y: 0.234375 },
  { x: 0.515625, y: 0.234375 },
  { x: 0.515625, y: 0.234375 },
  { x: 0.546875, y: 0.234375 },
  { x: 0.546875, y: 0.234375 },
  { x: 0.578125, y: 0.234375 },
  { x: 0.578125, y: 0.234375 },
  { x: 0.609375, y: 0.234375 },
  { x: 0.609375, y: 0.234375 },
  { x: 0.640625, y: 0.234375 },
  { x: 0.640625, y: 0.234375 },
  { x: 0.671875, y: 0.234375 },
  { x: 0.671875, y: 0.234375 },
  { x: 0.703125, y: 0.234375 },
  { x: 0.703125, y: 0.234375 },
  { x: 0.734375, y: 0.234375 },
  { x: 0.734375, y: 0.234375 },
  { x: 0.765625, y: 0.234375 },
  { x: 0.765625, y: 0.234375 },
  { x: 0.796875, y: 0.234375 },
  { x: 0.796875, y: 0.234375 },
  { x: 0.828125, y: 0.234375 },
  { x: 0.828125, y: 0.234375 },
  { x: 0.859375, y: 0.234375 },
  { x: 0.859375, y: 0.234375 },
  { x: 0.890625, y: 0.234375 },
  { x: 0.890625, y: 0.234375 },
  { x: 0.921875, y: 0.234375 },
  { x: 0.921875, y: 0.234375 },
  { x: 0.953125, y: 0.234375 },
  { x: 0.953125, y: 0.234375 },
  { x: 0.984375, y: 0.234375 },
  { x: 0.984375, y: 0.234375 },
  { x: 0.015625, y: 0.265625 },
  { x: 0.015625, y: 0.265625 },
  { x: 0.046875, y: 0.265625 },
  { x: 0.046875, y: 0.265625 },
  { x: 0.078125, y: 0.265625 },
  { x: 0.078125, y: 0.265625 },
  { x: 0.109375, y: 0.265625 },
  { x: 0.109375, y: 0.265625 },
  { x: 0.140625, y: 0.265625 },
  { x: 0.140625, y: 0.265625 },
  { x: 0.171875, y: 0.265625 },
  { x: 0.171875, y: 0.265625 },
  { x: 0.203125, y: 0.265625 },
  { x: 0.203125, y: 0.265625 },
  { x: 0.234375, y: 0.265625 },
  { x: 0.234375, y: 0.265625 },
  { x: 0.265625, y: 0.265625 },
  { x: 0.265625, y: 0.265625 },
  { x: 0.296875, y: 0.265625 },
  { x: 0.296875, y: 0.265625 },
  { x: 0.328125, y: 0.265625 },
  { x: 0.328125, y: 0.265625 },
  { x: 0.359375, y: 0.265625 },
  { x: 0.359375, y: 0.265625 },
  { x: 0.390625, y: 0.265625 },
  { x: 0.390625, y: 0.265625 },
  { x: 0.421875, y: 0.265625 },
  { x: 0.421875, y: 0.265625 },
  { x: 0.453125, y: 0.265625 },
  { x: 0.453125, y: 0.265625 },
  { x: 0.484375, y: 0.265625 },
  { x: 0.484375, y: 0.265625 },
  { x: 0.515625, y: 0.265625 },
  { x: 0.515625, y: 0.265625 },
  { x: 0.546875, y: 0.265625 },
  { x: 0.546875, y: 0.265625 },
  { x: 0.578125, y: 0.265625 },
  { x: 0.578125, y: 0.265625 },
  { x: 0.609375, y: 0.265625 },
  { x: 0.609375, y: 0.265625 },
  { x: 0.640625, y: 0.265625 },
  { x: 0.640625, y: 0.265625 },
  { x: 0.671875, y: 0.265625 },
  { x: 0.671875, y: 0.265625 },
  { x: 0.703125, y: 0.265625 },
  { x: 0.703125, y: 0.265625 },
  { x: 0.734375, y: 0.265625 },
  { x: 0.734375, y: 0.265625 },
  { x: 0.765625, y: 0.265625 },
  { x: 0.765625, y: 0.265625 },
  { x: 0.796875, y: 0.265625 },
  { x: 0.796875, y: 0.265625 },
  { x: 0.828125, y: 0.265625 },
  { x: 0.828125, y: 0.265625 },
  { x: 0.859375, y: 0.265625 },
  { x: 0.859375, y: 0.265625 },
  { x: 0.890625, y: 0.265625 },
  { x: 0.890625, y: 0.265625 },
  { x: 0.921875, y: 0.265625 },
  { x: 0.921875, y: 0.265625 },
  { x: 0.953125, y: 0.265625 },
  { x: 0.953125, y: 0.265625 },
  { x: 0.984375, y: 0.265625 },
  { x: 0.984375, y: 0.265625 },
  { x: 0.015625, y: 0.296875 },
  { x: 0.015625, y: 0.296875 },
  { x: 0.046875, y: 0.296875 },
  { x: 0.046875, y: 0.296875 },
  { x: 0.078125, y: 0.296875 },
  { x: 0.078125, y: 0.296875 },
  { x: 0.109375, y: 0.296875 },
  { x: 0.109375, y: 0.296875 },
  { x: 0.140625, y: 0.296875 },
  { x: 0.140625, y: 0.296875 },
  { x: 0.171875, y: 0.296875 },
  { x: 0.171875, y: 0.296875 },
  { x: 0.203125, y: 0.296875 },
  { x: 0.203125, y: 0.296875 },
  { x: 0.234375, y: 0.296875 },
  { x: 0.234375, y: 0.296875 },
  { x: 0.265625, y: 0.296875 },
  { x: 0.265625, y: 0.296875 },
  { x: 0.296875, y: 0.296875 },
  { x: 0.296875, y: 0.296875 },
  { x: 0.328125, y: 0.296875 },
  { x: 0.328125, y: 0.296875 },
  { x: 0.359375, y: 0.296875 },
  { x: 0.359375, y: 0.296875 },
  { x: 0.390625, y: 0.296875 },
  { x: 0.390625, y: 0.296875 },
  { x: 0.421875, y: 0.296875 },
  { x: 0.421875, y: 0.296875 },
  { x: 0.453125, y: 0.296875 },
  { x: 0.453125, y: 0.296875 },
  { x: 0.484375, y: 0.296875 },
  { x: 0.484375, y: 0.296875 },
  { x: 0.515625, y: 0.296875 },
  { x: 0.515625, y: 0.296875 },
  { x: 0.546875, y: 0.296875 },
  { x: 0.546875, y: 0.296875 },
  { x: 0.578125, y: 0.296875 },
  { x: 0.578125, y: 0.296875 },
  { x: 0.609375, y: 0.296875 },
  { x: 0.609375, y: 0.296875 },
  { x: 0.640625, y: 0.296875 },
  { x: 0.640625, y: 0.296875 },
  { x: 0.671875, y: 0.296875 },
  { x: 0.671875, y: 0.296875 },
  { x: 0.703125, y: 0.296875 },
  { x: 0.703125, y: 0.296875 },
  { x: 0.734375, y: 0.296875 },
  { x: 0.734375, y: 0.296875 },
  { x: 0.765625, y: 0.296875 },
  { x: 0.765625, y: 0.296875 },
  { x: 0.796875, y: 0.296875 },
  { x: 0.796875, y: 0.296875 },
  { x: 0.828125, y: 0.296875 },
  { x: 0.828125, y: 0.296875 },
  { x: 0.859375, y: 0.296875 },
  { x: 0.859375, y: 0.296875 },
  { x: 0.890625, y: 0.296875 },
  { x: 0.890625, y: 0.296875 },
  { x: 0.921875, y: 0.296875 },
  { x: 0.921875, y: 0.296875 },
  { x: 0.953125, y: 0.296875 },
  { x: 0.953125, y: 0.296875 },
  { x: 0.984375, y: 0.296875 },
  { x: 0.984375, y: 0.296875 },
  { x: 0.015625, y: 0.328125 },
  { x: 0.015625, y: 0.328125 },
  { x: 0.046875, y: 0.328125 },
  { x: 0.046875, y: 0.328125 },
  { x: 0.078125, y: 0.328125 },
  { x: 0.078125, y: 0.328125 },
  { x: 0.109375, y: 0.328125 },
  { x: 0.109375, y: 0.328125 },
  { x: 0.140625, y: 0.328125 },
  { x: 0.140625, y: 0.328125 },
  { x: 0.171875, y: 0.328125 },
  { x: 0.171875, y: 0.328125 },
  { x: 0.203125, y: 0.328125 },
  { x: 0.203125, y: 0.328125 },
  { x: 0.234375, y: 0.328125 },
  { x: 0.234375, y: 0.328125 },
  { x: 0.265625, y: 0.328125 },
  { x: 0.265625, y: 0.328125 },
  { x: 0.296875, y: 0.328125 },
  { x: 0.296875, y: 0.328125 },
  { x: 0.328125, y: 0.328125 },
  { x: 0.328125, y: 0.328125 },
  { x: 0.359375, y: 0.328125 },
  { x: 0.359375, y: 0.328125 },
  { x: 0.390625, y: 0.328125 },
  { x: 0.390625, y: 0.328125 },
  { x: 0.421875, y: 0.328125 },
  { x: 0.421875, y: 0.328125 },
  { x: 0.453125, y: 0.328125 },
  { x: 0.453125, y: 0.328125 },
  { x: 0.484375, y: 0.328125 },
  { x: 0.484375, y: 0.328125 },
  { x: 0.515625, y: 0.328125 },
  { x: 0.515625, y: 0.328125 },
  { x: 0.546875, y: 0.328125 },
  { x: 0.546875, y: 0.328125 },
  { x: 0.578125, y: 0.328125 },
  { x: 0.578125, y: 0.328125 },
  { x: 0.609375, y: 0.328125 },
  { x: 0.609375, y: 0.328125 },
  { x: 0.640625, y: 0.328125 },
  { x: 0.640625, y: 0.328125 },
  { x: 0.671875, y: 0.328125 },
  { x: 0.671875, y: 0.328125 },
  { x: 0.703125, y: 0.328125 },
  { x: 0.703125, y: 0.328125 },
  { x: 0.734375, y: 0.328125 },
  { x: 0.734375, y: 0.328125 },
  { x: 0.765625, y: 0.328125 },
  { x: 0.765625, y: 0.328125 },
  { x: 0.796875, y: 0.328125 },
  { x: 0.796875, y: 0.328125 },
  { x: 0.828125, y: 0.328125 },
  { x: 0.828125, y: 0.328125 },
  { x: 0.859375, y: 0.328125 },
  { x: 0.859375, y: 0.328125 },
  { x: 0.890625, y: 0.328125 },
  { x: 0.890625, y: 0.328125 },
  { x: 0.921875, y: 0.328125 },
  { x: 0.921875, y: 0.328125 },
  { x: 0.953125, y: 0.328125 },
  { x: 0.953125, y: 0.328125 },
  { x: 0.984375, y: 0.328125 },
  { x: 0.984375, y: 0.328125 },
  { x: 0.015625, y: 0.359375 },
  { x: 0.015625, y: 0.359375 },
  { x: 0.046875, y: 0.359375 },
  { x: 0.046875, y: 0.359375 },
  { x: 0.078125, y: 0.359375 },
  { x: 0.078125, y: 0.359375 },
  { x: 0.109375, y: 0.359375 },
  { x: 0.109375, y: 0.359375 },
  { x: 0.140625, y: 0.359375 },
  { x: 0.140625, y: 0.359375 },
  { x: 0.171875, y: 0.359375 },
  { x: 0.171875, y: 0.359375 },
  { x: 0.203125, y: 0.359375 },
  { x: 0.203125, y: 0.359375 },
  { x: 0.234375, y: 0.359375 },
  { x: 0.234375, y: 0.359375 },
  { x: 0.265625, y: 0.359375 },
  { x: 0.265625, y: 0.359375 },
  { x: 0.296875, y: 0.359375 },
  { x: 0.296875, y: 0.359375 },
  { x: 0.328125, y: 0.359375 },
  { x: 0.328125, y: 0.359375 },
  { x: 0.359375, y: 0.359375 },
  { x: 0.359375, y: 0.359375 },
  { x: 0.390625, y: 0.359375 },
  { x: 0.390625, y: 0.359375 },
  { x: 0.421875, y: 0.359375 },
  { x: 0.421875, y: 0.359375 },
  { x: 0.453125, y: 0.359375 },
  { x: 0.453125, y: 0.359375 },
  { x: 0.484375, y: 0.359375 },
  { x: 0.484375, y: 0.359375 },
  { x: 0.515625, y: 0.359375 },
  { x: 0.515625, y: 0.359375 },
  { x: 0.546875, y: 0.359375 },
  { x: 0.546875, y: 0.359375 },
  { x: 0.578125, y: 0.359375 },
  { x: 0.578125, y: 0.359375 },
  { x: 0.609375, y: 0.359375 },
  { x: 0.609375, y: 0.359375 },
  { x: 0.640625, y: 0.359375 },
  { x: 0.640625, y: 0.359375 },
  { x: 0.671875, y: 0.359375 },
  { x: 0.671875, y: 0.359375 },
  { x: 0.703125, y: 0.359375 },
  { x: 0.703125, y: 0.359375 },
  { x: 0.734375, y: 0.359375 },
  { x: 0.734375, y: 0.359375 },
  { x: 0.765625, y: 0.359375 },
  { x: 0.765625, y: 0.359375 },
  { x: 0.796875, y: 0.359375 },
  { x: 0.796875, y: 0.359375 },
  { x: 0.828125, y: 0.359375 },
  { x: 0.828125, y: 0.359375 },
  { x: 0.859375, y: 0.359375 },
  { x: 0.859375, y: 0.359375 },
  { x: 0.890625, y: 0.359375 },
  { x: 0.890625, y: 0.359375 },
  { x: 0.921875, y: 0.359375 },
  { x: 0.921875, y: 0.359375 },
  { x: 0.953125, y: 0.359375 },
  { x: 0.953125, y: 0.359375 },
  { x: 0.984375, y: 0.359375 },
  { x: 0.984375, y: 0.359375 },
  { x: 0.015625, y: 0.390625 },
  { x: 0.015625, y: 0.390625 },
  { x: 0.046875, y: 0.390625 },
  { x: 0.046875, y: 0.390625 },
  { x: 0.078125, y: 0.390625 },
  { x: 0.078125, y: 0.390625 },
  { x: 0.109375, y: 0.390625 },
  { x: 0.109375, y: 0.390625 },
  { x: 0.140625, y: 0.390625 },
  { x: 0.140625, y: 0.390625 },
  { x: 0.171875, y: 0.390625 },
  { x: 0.171875, y: 0.390625 },
  { x: 0.203125, y: 0.390625 },
  { x: 0.203125, y: 0.390625 },
  { x: 0.234375, y: 0.390625 },
  { x: 0.234375, y: 0.390625 },
  { x: 0.265625, y: 0.390625 },
  { x: 0.265625, y: 0.390625 },
  { x: 0.296875, y: 0.390625 },
  { x: 0.296875, y: 0.390625 },
  { x: 0.328125, y: 0.390625 },
  { x: 0.328125, y: 0.390625 },
  { x: 0.359375, y: 0.390625 },
  { x: 0.359375, y: 0.390625 },
  { x: 0.390625, y: 0.390625 },
  { x: 0.390625, y: 0.390625 },
  { x: 0.421875, y: 0.390625 },
  { x: 0.421875, y: 0.390625 },
  { x: 0.453125, y: 0.390625 },
  { x: 0.453125, y: 0.390625 },
  { x: 0.484375, y: 0.390625 },
  { x: 0.484375, y: 0.390625 },
  { x: 0.515625, y: 0.390625 },
  { x: 0.515625, y: 0.390625 },
  { x: 0.546875, y: 0.390625 },
  { x: 0.546875, y: 0.390625 },
  { x: 0.578125, y: 0.390625 },
  { x: 0.578125, y: 0.390625 },
  { x: 0.609375, y: 0.390625 },
  { x: 0.609375, y: 0.390625 },
  { x: 0.640625, y: 0.390625 },
  { x: 0.640625, y: 0.390625 },
  { x: 0.671875, y: 0.390625 },
  { x: 0.671875, y: 0.390625 },
  { x: 0.703125, y: 0.390625 },
  { x: 0.703125, y: 0.390625 },
  { x: 0.734375, y: 0.390625 },
  { x: 0.734375, y: 0.390625 },
  { x: 0.765625, y: 0.390625 },
  { x: 0.765625, y: 0.390625 },
  { x: 0.796875, y: 0.390625 },
  { x: 0.796875, y: 0.390625 },
  { x: 0.828125, y: 0.390625 },
  { x: 0.828125, y: 0.390625 },
  { x: 0.859375, y: 0.390625 },
  { x: 0.859375, y: 0.390625 },
  { x: 0.890625, y: 0.390625 },
  { x: 0.890625, y: 0.390625 },
  { x: 0.921875, y: 0.390625 },
  { x: 0.921875, y: 0.390625 },
  { x: 0.953125, y: 0.390625 },
  { x: 0.953125, y: 0.390625 },
  { x: 0.984375, y: 0.390625 },
  { x: 0.984375, y: 0.390625 },
  { x: 0.015625, y: 0.421875 },
  { x: 0.015625, y: 0.421875 },
  { x: 0.046875, y: 0.421875 },
  { x: 0.046875, y: 0.421875 },
  { x: 0.078125, y: 0.421875 },
  { x: 0.078125, y: 0.421875 },
  { x: 0.109375, y: 0.421875 },
  { x: 0.109375, y: 0.421875 },
  { x: 0.140625, y: 0.421875 },
  { x: 0.140625, y: 0.421875 },
  { x: 0.171875, y: 0.421875 },
  { x: 0.171875, y: 0.421875 },
  { x: 0.203125, y: 0.421875 },
  { x: 0.203125, y: 0.421875 },
  { x: 0.234375, y: 0.421875 },
  { x: 0.234375, y: 0.421875 },
  { x: 0.265625, y: 0.421875 },
  { x: 0.265625, y: 0.421875 },
  { x: 0.296875, y: 0.421875 },
  { x: 0.296875, y: 0.421875 },
  { x: 0.328125, y: 0.421875 },
  { x: 0.328125, y: 0.421875 },
  { x: 0.359375, y: 0.421875 },
  { x: 0.359375, y: 0.421875 },
  { x: 0.390625, y: 0.421875 },
  { x: 0.390625, y: 0.421875 },
  { x: 0.421875, y: 0.421875 },
  { x: 0.421875, y: 0.421875 },
  { x: 0.453125, y: 0.421875 },
  { x: 0.453125, y: 0.421875 },
  { x: 0.484375, y: 0.421875 },
  { x: 0.484375, y: 0.421875 },
  { x: 0.515625, y: 0.421875 },
  { x: 0.515625, y: 0.421875 },
  { x: 0.546875, y: 0.421875 },
  { x: 0.546875, y: 0.421875 },
  { x: 0.578125, y: 0.421875 },
  { x: 0.578125, y: 0.421875 },
  { x: 0.609375, y: 0.421875 },
  { x: 0.609375, y: 0.421875 },
  { x: 0.640625, y: 0.421875 },
  { x: 0.640625, y: 0.421875 },
  { x: 0.671875, y: 0.421875 },
  { x: 0.671875, y: 0.421875 },
  { x: 0.703125, y: 0.421875 },
  { x: 0.703125, y: 0.421875 },
  { x: 0.734375, y: 0.421875 },
  { x: 0.734375, y: 0.421875 },
  { x: 0.765625, y: 0.421875 },
  { x: 0.765625, y: 0.421875 },
  { x: 0.796875, y: 0.421875 },
  { x: 0.796875, y: 0.421875 },
  { x: 0.828125, y: 0.421875 },
  { x: 0.828125, y: 0.421875 },
  { x: 0.859375, y: 0.421875 },
  { x: 0.859375, y: 0.421875 },
  { x: 0.890625, y: 0.421875 },
  { x: 0.890625, y: 0.421875 },
  { x: 0.921875, y: 0.421875 },
  { x: 0.921875, y: 0.421875 },
  { x: 0.953125, y: 0.421875 },
  { x: 0.953125, y: 0.421875 },
  { x: 0.984375, y: 0.421875 },
  { x: 0.984375, y: 0.421875 },
  { x: 0.015625, y: 0.453125 },
  { x: 0.015625, y: 0.453125 },
  { x: 0.046875, y: 0.453125 },
  { x: 0.046875, y: 0.453125 },
  { x: 0.078125, y: 0.453125 },
  { x: 0.078125, y: 0.453125 },
  { x: 0.109375, y: 0.453125 },
  { x: 0.109375, y: 0.453125 },
  { x: 0.140625, y: 0.453125 },
  { x: 0.140625, y: 0.453125 },
  { x: 0.171875, y: 0.453125 },
  { x: 0.171875, y: 0.453125 },
  { x: 0.203125, y: 0.453125 },
  { x: 0.203125, y: 0.453125 },
  { x: 0.234375, y: 0.453125 },
  { x: 0.234375, y: 0.453125 },
  { x: 0.265625, y: 0.453125 },
  { x: 0.265625, y: 0.453125 },
  { x: 0.296875, y: 0.453125 },
  { x: 0.296875, y: 0.453125 },
  { x: 0.328125, y: 0.453125 },
  { x: 0.328125, y: 0.453125 },
  { x: 0.359375, y: 0.453125 },
  { x: 0.359375, y: 0.453125 },
  { x: 0.390625, y: 0.453125 },
  { x: 0.390625, y: 0.453125 },
  { x: 0.421875, y: 0.453125 },
  { x: 0.421875, y: 0.453125 },
  { x: 0.453125, y: 0.453125 },
  { x: 0.453125, y: 0.453125 },
  { x: 0.484375, y: 0.453125 },
  { x: 0.484375, y: 0.453125 },
  { x: 0.515625, y: 0.453125 },
  { x: 0.515625, y: 0.453125 },
  { x: 0.546875, y: 0.453125 },
  { x: 0.546875, y: 0.453125 },
  { x: 0.578125, y: 0.453125 },
  { x: 0.578125, y: 0.453125 },
  { x: 0.609375, y: 0.453125 },
  { x: 0.609375, y: 0.453125 },
  { x: 0.640625, y: 0.453125 },
  { x: 0.640625, y: 0.453125 },
  { x: 0.671875, y: 0.453125 },
  { x: 0.671875, y: 0.453125 },
  { x: 0.703125, y: 0.453125 },
  { x: 0.703125, y: 0.453125 },
  { x: 0.734375, y: 0.453125 },
  { x: 0.734375, y: 0.453125 },
  { x: 0.765625, y: 0.453125 },
  { x: 0.765625, y: 0.453125 },
  { x: 0.796875, y: 0.453125 },
  { x: 0.796875, y: 0.453125 },
  { x: 0.828125, y: 0.453125 },
  { x: 0.828125, y: 0.453125 },
  { x: 0.859375, y: 0.453125 },
  { x: 0.859375, y: 0.453125 },
  { x: 0.890625, y: 0.453125 },
  { x: 0.890625, y: 0.453125 },
  { x: 0.921875, y: 0.453125 },
  { x: 0.921875, y: 0.453125 },
  { x: 0.953125, y: 0.453125 },
  { x: 0.953125, y: 0.453125 },
  { x: 0.984375, y: 0.453125 },
  { x: 0.984375, y: 0.453125 },
  { x: 0.015625, y: 0.484375 },
  { x: 0.015625, y: 0.484375 },
  { x: 0.046875, y: 0.484375 },
  { x: 0.046875, y: 0.484375 },
  { x: 0.078125, y: 0.484375 },
  { x: 0.078125, y: 0.484375 },
  { x: 0.109375, y: 0.484375 },
  { x: 0.109375, y: 0.484375 },
  { x: 0.140625, y: 0.484375 },
  { x: 0.140625, y: 0.484375 },
  { x: 0.171875, y: 0.484375 },
  { x: 0.171875, y: 0.484375 },
  { x: 0.203125, y: 0.484375 },
  { x: 0.203125, y: 0.484375 },
  { x: 0.234375, y: 0.484375 },
  { x: 0.234375, y: 0.484375 },
  { x: 0.265625, y: 0.484375 },
  { x: 0.265625, y: 0.484375 },
  { x: 0.296875, y: 0.484375 },
  { x: 0.296875, y: 0.484375 },
  { x: 0.328125, y: 0.484375 },
  { x: 0.328125, y: 0.484375 },
  { x: 0.359375, y: 0.484375 },
  { x: 0.359375, y: 0.484375 },
  { x: 0.390625, y: 0.484375 },
  { x: 0.390625, y: 0.484375 },
  { x: 0.421875, y: 0.484375 },
  { x: 0.421875, y: 0.484375 },
  { x: 0.453125, y: 0.484375 },
  { x: 0.453125, y: 0.484375 },
  { x: 0.484375, y: 0.484375 },
  { x: 0.484375, y: 0.484375 },
  { x: 0.515625, y: 0.484375 },
  { x: 0.515625, y: 0.484375 },
  { x: 0.546875, y: 0.484375 },
  { x: 0.546875, y: 0.484375 },
  { x: 0.578125, y: 0.484375 },
  { x: 0.578125, y: 0.484375 },
  { x: 0.609375, y: 0.484375 },
  { x: 0.609375, y: 0.484375 },
  { x: 0.640625, y: 0.484375 },
  { x: 0.640625, y: 0.484375 },
  { x: 0.671875, y: 0.484375 },
  { x: 0.671875, y: 0.484375 },
  { x: 0.703125, y: 0.484375 },
  { x: 0.703125, y: 0.484375 },
  { x: 0.734375, y: 0.484375 },
  { x: 0.734375, y: 0.484375 },
  { x: 0.765625, y: 0.484375 },
  { x: 0.765625, y: 0.484375 },
  { x: 0.796875, y: 0.484375 },
  { x: 0.796875, y: 0.484375 },
  { x: 0.828125, y: 0.484375 },
  { x: 0.828125, y: 0.484375 },
  { x: 0.859375, y: 0.484375 },
  { x: 0.859375, y: 0.484375 },
  { x: 0.890625, y: 0.484375 },
  { x: 0.890625, y: 0.484375 },
  { x: 0.921875, y: 0.484375 },
  { x: 0.921875, y: 0.484375 },
  { x: 0.953125, y: 0.484375 },
  { x: 0.953125, y: 0.484375 },
  { x: 0.984375, y: 0.484375 },
  { x: 0.984375, y: 0.484375 },
  { x: 0.015625, y: 0.515625 },
  { x: 0.015625, y: 0.515625 },
  { x: 0.046875, y: 0.515625 },
  { x: 0.046875, y: 0.515625 },
  { x: 0.078125, y: 0.515625 },
  { x: 0.078125, y: 0.515625 },
  { x: 0.109375, y: 0.515625 },
  { x: 0.109375, y: 0.515625 },
  { x: 0.140625, y: 0.515625 },
  { x: 0.140625, y: 0.515625 },
  { x: 0.171875, y: 0.515625 },
  { x: 0.171875, y: 0.515625 },
  { x: 0.203125, y: 0.515625 },
  { x: 0.203125, y: 0.515625 },
  { x: 0.234375, y: 0.515625 },
  { x: 0.234375, y: 0.515625 },
  { x: 0.265625, y: 0.515625 },
  { x: 0.265625, y: 0.515625 },
  { x: 0.296875, y: 0.515625 },
  { x: 0.296875, y: 0.515625 },
  { x: 0.328125, y: 0.515625 },
  { x: 0.328125, y: 0.515625 },
  { x: 0.359375, y: 0.515625 },
  { x: 0.359375, y: 0.515625 },
  { x: 0.390625, y: 0.515625 },
  { x: 0.390625, y: 0.515625 },
  { x: 0.421875, y: 0.515625 },
  { x: 0.421875, y: 0.515625 },
  { x: 0.453125, y: 0.515625 },
  { x: 0.453125, y: 0.515625 },
  { x: 0.484375, y: 0.515625 },
  { x: 0.484375, y: 0.515625 },
  { x: 0.515625, y: 0.515625 },
  { x: 0.515625, y: 0.515625 },
  { x: 0.546875, y: 0.515625 },
  { x: 0.546875, y: 0.515625 },
  { x: 0.578125, y: 0.515625 },
  { x: 0.578125, y: 0.515625 },
  { x: 0.609375, y: 0.515625 },
  { x: 0.609375, y: 0.515625 },
  { x: 0.640625, y: 0.515625 },
  { x: 0.640625, y: 0.515625 },
  { x: 0.671875, y: 0.515625 },
  { x: 0.671875, y: 0.515625 },
  { x: 0.703125, y: 0.515625 },
  { x: 0.703125, y: 0.515625 },
  { x: 0.734375, y: 0.515625 },
  { x: 0.734375, y: 0.515625 },
  { x: 0.765625, y: 0.515625 },
  { x: 0.765625, y: 0.515625 },
  { x: 0.796875, y: 0.515625 },
  { x: 0.796875, y: 0.515625 },
  { x: 0.828125, y: 0.515625 },
  { x: 0.828125, y: 0.515625 },
  { x: 0.859375, y: 0.515625 },
  { x: 0.859375, y: 0.515625 },
  { x: 0.890625, y: 0.515625 },
  { x: 0.890625, y: 0.515625 },
  { x: 0.921875, y: 0.515625 },
  { x: 0.921875, y: 0.515625 },
  { x: 0.953125, y: 0.515625 },
  { x: 0.953125, y: 0.515625 },
  { x: 0.984375, y: 0.515625 },
  { x: 0.984375, y: 0.515625 },
  { x: 0.015625, y: 0.546875 },
  { x: 0.015625, y: 0.546875 },
  { x: 0.046875, y: 0.546875 },
  { x: 0.046875, y: 0.546875 },
  { x: 0.078125, y: 0.546875 },
  { x: 0.078125, y: 0.546875 },
  { x: 0.109375, y: 0.546875 },
  { x: 0.109375, y: 0.546875 },
  { x: 0.140625, y: 0.546875 },
  { x: 0.140625, y: 0.546875 },
  { x: 0.171875, y: 0.546875 },
  { x: 0.171875, y: 0.546875 },
  { x: 0.203125, y: 0.546875 },
  { x: 0.203125, y: 0.546875 },
  { x: 0.234375, y: 0.546875 },
  { x: 0.234375, y: 0.546875 },
  { x: 0.265625, y: 0.546875 },
  { x: 0.265625, y: 0.546875 },
  { x: 0.296875, y: 0.546875 },
  { x: 0.296875, y: 0.546875 },
  { x: 0.328125, y: 0.546875 },
  { x: 0.328125, y: 0.546875 },
  { x: 0.359375, y: 0.546875 },
  { x: 0.359375, y: 0.546875 },
  { x: 0.390625, y: 0.546875 },
  { x: 0.390625, y: 0.546875 },
  { x: 0.421875, y: 0.546875 },
  { x: 0.421875, y: 0.546875 },
  { x: 0.453125, y: 0.546875 },
  { x: 0.453125, y: 0.546875 },
  { x: 0.484375, y: 0.546875 },
  { x: 0.484375, y: 0.546875 },
  { x: 0.515625, y: 0.546875 },
  { x: 0.515625, y: 0.546875 },
  { x: 0.546875, y: 0.546875 },
  { x: 0.546875, y: 0.546875 },
  { x: 0.578125, y: 0.546875 },
  { x: 0.578125, y: 0.546875 },
  { x: 0.609375, y: 0.546875 },
  { x: 0.609375, y: 0.546875 },
  { x: 0.640625, y: 0.546875 },
  { x: 0.640625, y: 0.546875 },
  { x: 0.671875, y: 0.546875 },
  { x: 0.671875, y: 0.546875 },
  { x: 0.703125, y: 0.546875 },
  { x: 0.703125, y: 0.546875 },
  { x: 0.734375, y: 0.546875 },
  { x: 0.734375, y: 0.546875 },
  { x: 0.765625, y: 0.546875 },
  { x: 0.765625, y: 0.546875 },
  { x: 0.796875, y: 0.546875 },
  { x: 0.796875, y: 0.546875 },
  { x: 0.828125, y: 0.546875 },
  { x: 0.828125, y: 0.546875 },
  { x: 0.859375, y: 0.546875 },
  { x: 0.859375, y: 0.546875 },
  { x: 0.890625, y: 0.546875 },
  { x: 0.890625, y: 0.546875 },
  { x: 0.921875, y: 0.546875 },
  { x: 0.921875, y: 0.546875 },
  { x: 0.953125, y: 0.546875 },
  { x: 0.953125, y: 0.546875 },
  { x: 0.984375, y: 0.546875 },
  { x: 0.984375, y: 0.546875 },
  { x: 0.015625, y: 0.578125 },
  { x: 0.015625, y: 0.578125 },
  { x: 0.046875, y: 0.578125 },
  { x: 0.046875, y: 0.578125 },
  { x: 0.078125, y: 0.578125 },
  { x: 0.078125, y: 0.578125 },
  { x: 0.109375, y: 0.578125 },
  { x: 0.109375, y: 0.578125 },
  { x: 0.140625, y: 0.578125 },
  { x: 0.140625, y: 0.578125 },
  { x: 0.171875, y: 0.578125 },
  { x: 0.171875, y: 0.578125 },
  { x: 0.203125, y: 0.578125 },
  { x: 0.203125, y: 0.578125 },
  { x: 0.234375, y: 0.578125 },
  { x: 0.234375, y: 0.578125 },
  { x: 0.265625, y: 0.578125 },
  { x: 0.265625, y: 0.578125 },
  { x: 0.296875, y: 0.578125 },
  { x: 0.296875, y: 0.578125 },
  { x: 0.328125, y: 0.578125 },
  { x: 0.328125, y: 0.578125 },
  { x: 0.359375, y: 0.578125 },
  { x: 0.359375, y: 0.578125 },
  { x: 0.390625, y: 0.578125 },
  { x: 0.390625, y: 0.578125 },
  { x: 0.421875, y: 0.578125 },
  { x: 0.421875, y: 0.578125 },
  { x: 0.453125, y: 0.578125 },
  { x: 0.453125, y: 0.578125 },
  { x: 0.484375, y: 0.578125 },
  { x: 0.484375, y: 0.578125 },
  { x: 0.515625, y: 0.578125 },
  { x: 0.515625, y: 0.578125 },
  { x: 0.546875, y: 0.578125 },
  { x: 0.546875, y: 0.578125 },
  { x: 0.578125, y: 0.578125 },
  { x: 0.578125, y: 0.578125 },
  { x: 0.609375, y: 0.578125 },
  { x: 0.609375, y: 0.578125 },
  { x: 0.640625, y: 0.578125 },
  { x: 0.640625, y: 0.578125 },
  { x: 0.671875, y: 0.578125 },
  { x: 0.671875, y: 0.578125 },
  { x: 0.703125, y: 0.578125 },
  { x: 0.703125, y: 0.578125 },
  { x: 0.734375, y: 0.578125 },
  { x: 0.734375, y: 0.578125 },
  { x: 0.765625, y: 0.578125 },
  { x: 0.765625, y: 0.578125 },
  { x: 0.796875, y: 0.578125 },
  { x: 0.796875, y: 0.578125 },
  { x: 0.828125, y: 0.578125 },
  { x: 0.828125, y: 0.578125 },
  { x: 0.859375, y: 0.578125 },
  { x: 0.859375, y: 0.578125 },
  { x: 0.890625, y: 0.578125 },
  { x: 0.890625, y: 0.578125 },
  { x: 0.921875, y: 0.578125 },
  { x: 0.921875, y: 0.578125 },
  { x: 0.953125, y: 0.578125 },
  { x: 0.953125, y: 0.578125 },
  { x: 0.984375, y: 0.578125 },
  { x: 0.984375, y: 0.578125 },
  { x: 0.015625, y: 0.609375 },
  { x: 0.015625, y: 0.609375 },
  { x: 0.046875, y: 0.609375 },
  { x: 0.046875, y: 0.609375 },
  { x: 0.078125, y: 0.609375 },
  { x: 0.078125, y: 0.609375 },
  { x: 0.109375, y: 0.609375 },
  { x: 0.109375, y: 0.609375 },
  { x: 0.140625, y: 0.609375 },
  { x: 0.140625, y: 0.609375 },
  { x: 0.171875, y: 0.609375 },
  { x: 0.171875, y: 0.609375 },
  { x: 0.203125, y: 0.609375 },
  { x: 0.203125, y: 0.609375 },
  { x: 0.234375, y: 0.609375 },
  { x: 0.234375, y: 0.609375 },
  { x: 0.265625, y: 0.609375 },
  { x: 0.265625, y: 0.609375 },
  { x: 0.296875, y: 0.609375 },
  { x: 0.296875, y: 0.609375 },
  { x: 0.328125, y: 0.609375 },
  { x: 0.328125, y: 0.609375 },
  { x: 0.359375, y: 0.609375 },
  { x: 0.359375, y: 0.609375 },
  { x: 0.390625, y: 0.609375 },
  { x: 0.390625, y: 0.609375 },
  { x: 0.421875, y: 0.609375 },
  { x: 0.421875, y: 0.609375 },
  { x: 0.453125, y: 0.609375 },
  { x: 0.453125, y: 0.609375 },
  { x: 0.484375, y: 0.609375 },
  { x: 0.484375, y: 0.609375 },
  { x: 0.515625, y: 0.609375 },
  { x: 0.515625, y: 0.609375 },
  { x: 0.546875, y: 0.609375 },
  { x: 0.546875, y: 0.609375 },
  { x: 0.578125, y: 0.609375 },
  { x: 0.578125, y: 0.609375 },
  { x: 0.609375, y: 0.609375 },
  { x: 0.609375, y: 0.609375 },
  { x: 0.640625, y: 0.609375 },
  { x: 0.640625, y: 0.609375 },
  { x: 0.671875, y: 0.609375 },
  { x: 0.671875, y: 0.609375 },
  { x: 0.703125, y: 0.609375 },
  { x: 0.703125, y: 0.609375 },
  { x: 0.734375, y: 0.609375 },
  { x: 0.734375, y: 0.609375 },
  { x: 0.765625, y: 0.609375 },
  { x: 0.765625, y: 0.609375 },
  { x: 0.796875, y: 0.609375 },
  { x: 0.796875, y: 0.609375 },
  { x: 0.828125, y: 0.609375 },
  { x: 0.828125, y: 0.609375 },
  { x: 0.859375, y: 0.609375 },
  { x: 0.859375, y: 0.609375 },
  { x: 0.890625, y: 0.609375 },
  { x: 0.890625, y: 0.609375 },
  { x: 0.921875, y: 0.609375 },
  { x: 0.921875, y: 0.609375 },
  { x: 0.953125, y: 0.609375 },
  { x: 0.953125, y: 0.609375 },
  { x: 0.984375, y: 0.609375 },
  { x: 0.984375, y: 0.609375 },
  { x: 0.015625, y: 0.640625 },
  { x: 0.015625, y: 0.640625 },
  { x: 0.046875, y: 0.640625 },
  { x: 0.046875, y: 0.640625 },
  { x: 0.078125, y: 0.640625 },
  { x: 0.078125, y: 0.640625 },
  { x: 0.109375, y: 0.640625 },
  { x: 0.109375, y: 0.640625 },
  { x: 0.140625, y: 0.640625 },
  { x: 0.140625, y: 0.640625 },
  { x: 0.171875, y: 0.640625 },
  { x: 0.171875, y: 0.640625 },
  { x: 0.203125, y: 0.640625 },
  { x: 0.203125, y: 0.640625 },
  { x: 0.234375, y: 0.640625 },
  { x: 0.234375, y: 0.640625 },
  { x: 0.265625, y: 0.640625 },
  { x: 0.265625, y: 0.640625 },
  { x: 0.296875, y: 0.640625 },
  { x: 0.296875, y: 0.640625 },
  { x: 0.328125, y: 0.640625 },
  { x: 0.328125, y: 0.640625 },
  { x: 0.359375, y: 0.640625 },
  { x: 0.359375, y: 0.640625 },
  { x: 0.390625, y: 0.640625 },
  { x: 0.390625, y: 0.640625 },
  { x: 0.421875, y: 0.640625 },
  { x: 0.421875, y: 0.640625 },
  { x: 0.453125, y: 0.640625 },
  { x: 0.453125, y: 0.640625 },
  { x: 0.484375, y: 0.640625 },
  { x: 0.484375, y: 0.640625 },
  { x: 0.515625, y: 0.640625 },
  { x: 0.515625, y: 0.640625 },
  { x: 0.546875, y: 0.640625 },
  { x: 0.546875, y: 0.640625 },
  { x: 0.578125, y: 0.640625 },
  { x: 0.578125, y: 0.640625 },
  { x: 0.609375, y: 0.640625 },
  { x: 0.609375, y: 0.640625 },
  { x: 0.640625, y: 0.640625 },
  { x: 0.640625, y: 0.640625 },
  { x: 0.671875, y: 0.640625 },
  { x: 0.671875, y: 0.640625 },
  { x: 0.703125, y: 0.640625 },
  { x: 0.703125, y: 0.640625 },
  { x: 0.734375, y: 0.640625 },
  { x: 0.734375, y: 0.640625 },
  { x: 0.765625, y: 0.640625 },
  { x: 0.765625, y: 0.640625 },
  { x: 0.796875, y: 0.640625 },
  { x: 0.796875, y: 0.640625 },
  { x: 0.828125, y: 0.640625 },
  { x: 0.828125, y: 0.640625 },
  { x: 0.859375, y: 0.640625 },
  { x: 0.859375, y: 0.640625 },
  { x: 0.890625, y: 0.640625 },
  { x: 0.890625, y: 0.640625 },
  { x: 0.921875, y: 0.640625 },
  { x: 0.921875, y: 0.640625 },
  { x: 0.953125, y: 0.640625 },
  { x: 0.953125, y: 0.640625 },
  { x: 0.984375, y: 0.640625 },
  { x: 0.984375, y: 0.640625 },
  { x: 0.015625, y: 0.671875 },
  { x: 0.015625, y: 0.671875 },
  { x: 0.046875, y: 0.671875 },
  { x: 0.046875, y: 0.671875 },
  { x: 0.078125, y: 0.671875 },
  { x: 0.078125, y: 0.671875 },
  { x: 0.109375, y: 0.671875 },
  { x: 0.109375, y: 0.671875 },
  { x: 0.140625, y: 0.671875 },
  { x: 0.140625, y: 0.671875 },
  { x: 0.171875, y: 0.671875 },
  { x: 0.171875, y: 0.671875 },
  { x: 0.203125, y: 0.671875 },
  { x: 0.203125, y: 0.671875 },
  { x: 0.234375, y: 0.671875 },
  { x: 0.234375, y: 0.671875 },
  { x: 0.265625, y: 0.671875 },
  { x: 0.265625, y: 0.671875 },
  { x: 0.296875, y: 0.671875 },
  { x: 0.296875, y: 0.671875 },
  { x: 0.328125, y: 0.671875 },
  { x: 0.328125, y: 0.671875 },
  { x: 0.359375, y: 0.671875 },
  { x: 0.359375, y: 0.671875 },
  { x: 0.390625, y: 0.671875 },
  { x: 0.390625, y: 0.671875 },
  { x: 0.421875, y: 0.671875 },
  { x: 0.421875, y: 0.671875 },
  { x: 0.453125, y: 0.671875 },
  { x: 0.453125, y: 0.671875 },
  { x: 0.484375, y: 0.671875 },
  { x: 0.484375, y: 0.671875 },
  { x: 0.515625, y: 0.671875 },
  { x: 0.515625, y: 0.671875 },
  { x: 0.546875, y: 0.671875 },
  { x: 0.546875, y: 0.671875 },
  { x: 0.578125, y: 0.671875 },
  { x: 0.578125, y: 0.671875 },
  { x: 0.609375, y: 0.671875 },
  { x: 0.609375, y: 0.671875 },
  { x: 0.640625, y: 0.671875 },
  { x: 0.640625, y: 0.671875 },
  { x: 0.671875, y: 0.671875 },
  { x: 0.671875, y: 0.671875 },
  { x: 0.703125, y: 0.671875 },
  { x: 0.703125, y: 0.671875 },
  { x: 0.734375, y: 0.671875 },
  { x: 0.734375, y: 0.671875 },
  { x: 0.765625, y: 0.671875 },
  { x: 0.765625, y: 0.671875 },
  { x: 0.796875, y: 0.671875 },
  { x: 0.796875, y: 0.671875 },
  { x: 0.828125, y: 0.671875 },
  { x: 0.828125, y: 0.671875 },
  { x: 0.859375, y: 0.671875 },
  { x: 0.859375, y: 0.671875 },
  { x: 0.890625, y: 0.671875 },
  { x: 0.890625, y: 0.671875 },
  { x: 0.921875, y: 0.671875 },
  { x: 0.921875, y: 0.671875 },
  { x: 0.953125, y: 0.671875 },
  { x: 0.953125, y: 0.671875 },
  { x: 0.984375, y: 0.671875 },
  { x: 0.984375, y: 0.671875 },
  { x: 0.015625, y: 0.703125 },
  { x: 0.015625, y: 0.703125 },
  { x: 0.046875, y: 0.703125 },
  { x: 0.046875, y: 0.703125 },
  { x: 0.078125, y: 0.703125 },
  { x: 0.078125, y: 0.703125 },
  { x: 0.109375, y: 0.703125 },
  { x: 0.109375, y: 0.703125 },
  { x: 0.140625, y: 0.703125 },
  { x: 0.140625, y: 0.703125 },
  { x: 0.171875, y: 0.703125 },
  { x: 0.171875, y: 0.703125 },
  { x: 0.203125, y: 0.703125 },
  { x: 0.203125, y: 0.703125 },
  { x: 0.234375, y: 0.703125 },
  { x: 0.234375, y: 0.703125 },
  { x: 0.265625, y: 0.703125 },
  { x: 0.265625, y: 0.703125 },
  { x: 0.296875, y: 0.703125 },
  { x: 0.296875, y: 0.703125 },
  { x: 0.328125, y: 0.703125 },
  { x: 0.328125, y: 0.703125 },
  { x: 0.359375, y: 0.703125 },
  { x: 0.359375, y: 0.703125 },
  { x: 0.390625, y: 0.703125 },
  { x: 0.390625, y: 0.703125 },
  { x: 0.421875, y: 0.703125 },
  { x: 0.421875, y: 0.703125 },
  { x: 0.453125, y: 0.703125 },
  { x: 0.453125, y: 0.703125 },
  { x: 0.484375, y: 0.703125 },
  { x: 0.484375, y: 0.703125 },
  { x: 0.515625, y: 0.703125 },
  { x: 0.515625, y: 0.703125 },
  { x: 0.546875, y: 0.703125 },
  { x: 0.546875, y: 0.703125 },
  { x: 0.578125, y: 0.703125 },
  { x: 0.578125, y: 0.703125 },
  { x: 0.609375, y: 0.703125 },
  { x: 0.609375, y: 0.703125 },
  { x: 0.640625, y: 0.703125 },
  { x: 0.640625, y: 0.703125 },
  { x: 0.671875, y: 0.703125 },
  { x: 0.671875, y: 0.703125 },
  { x: 0.703125, y: 0.703125 },
  { x: 0.703125, y: 0.703125 },
  { x: 0.734375, y: 0.703125 },
  { x: 0.734375, y: 0.703125 },
  { x: 0.765625, y: 0.703125 },
  { x: 0.765625, y: 0.703125 },
  { x: 0.796875, y: 0.703125 },
  { x: 0.796875, y: 0.703125 },
  { x: 0.828125, y: 0.703125 },
  { x: 0.828125, y: 0.703125 },
  { x: 0.859375, y: 0.703125 },
  { x: 0.859375, y: 0.703125 },
  { x: 0.890625, y: 0.703125 },
  { x: 0.890625, y: 0.703125 },
  { x: 0.921875, y: 0.703125 },
  { x: 0.921875, y: 0.703125 },
  { x: 0.953125, y: 0.703125 },
  { x: 0.953125, y: 0.703125 },
  { x: 0.984375, y: 0.703125 },
  { x: 0.984375, y: 0.703125 },
  { x: 0.015625, y: 0.734375 },
  { x: 0.015625, y: 0.734375 },
  { x: 0.046875, y: 0.734375 },
  { x: 0.046875, y: 0.734375 },
  { x: 0.078125, y: 0.734375 },
  { x: 0.078125, y: 0.734375 },
  { x: 0.109375, y: 0.734375 },
  { x: 0.109375, y: 0.734375 },
  { x: 0.140625, y: 0.734375 },
  { x: 0.140625, y: 0.734375 },
  { x: 0.171875, y: 0.734375 },
  { x: 0.171875, y: 0.734375 },
  { x: 0.203125, y: 0.734375 },
  { x: 0.203125, y: 0.734375 },
  { x: 0.234375, y: 0.734375 },
  { x: 0.234375, y: 0.734375 },
  { x: 0.265625, y: 0.734375 },
  { x: 0.265625, y: 0.734375 },
  { x: 0.296875, y: 0.734375 },
  { x: 0.296875, y: 0.734375 },
  { x: 0.328125, y: 0.734375 },
  { x: 0.328125, y: 0.734375 },
  { x: 0.359375, y: 0.734375 },
  { x: 0.359375, y: 0.734375 },
  { x: 0.390625, y: 0.734375 },
  { x: 0.390625, y: 0.734375 },
  { x: 0.421875, y: 0.734375 },
  { x: 0.421875, y: 0.734375 },
  { x: 0.453125, y: 0.734375 },
  { x: 0.453125, y: 0.734375 },
  { x: 0.484375, y: 0.734375 },
  { x: 0.484375, y: 0.734375 },
  { x: 0.515625, y: 0.734375 },
  { x: 0.515625, y: 0.734375 },
  { x: 0.546875, y: 0.734375 },
  { x: 0.546875, y: 0.734375 },
  { x: 0.578125, y: 0.734375 },
  { x: 0.578125, y: 0.734375 },
  { x: 0.609375, y: 0.734375 },
  { x: 0.609375, y: 0.734375 },
  { x: 0.640625, y: 0.734375 },
  { x: 0.640625, y: 0.734375 },
  { x: 0.671875, y: 0.734375 },
  { x: 0.671875, y: 0.734375 },
  { x: 0.703125, y: 0.734375 },
  { x: 0.703125, y: 0.734375 },
  { x: 0.734375, y: 0.734375 },
  { x: 0.734375, y: 0.734375 },
  { x: 0.765625, y: 0.734375 },
  { x: 0.765625, y: 0.734375 },
  { x: 0.796875, y: 0.734375 },
  { x: 0.796875, y: 0.734375 },
  { x: 0.828125, y: 0.734375 },
  { x: 0.828125, y: 0.734375 },
  { x: 0.859375, y: 0.734375 },
  { x: 0.859375, y: 0.734375 },
  { x: 0.890625, y: 0.734375 },
  { x: 0.890625, y: 0.734375 },
  { x: 0.921875, y: 0.734375 },
  { x: 0.921875, y: 0.734375 },
  { x: 0.953125, y: 0.734375 },
  { x: 0.953125, y: 0.734375 },
  { x: 0.984375, y: 0.734375 },
  { x: 0.984375, y: 0.734375 },
  { x: 0.015625, y: 0.765625 },
  { x: 0.015625, y: 0.765625 },
  { x: 0.046875, y: 0.765625 },
  { x: 0.046875, y: 0.765625 },
  { x: 0.078125, y: 0.765625 },
  { x: 0.078125, y: 0.765625 },
  { x: 0.109375, y: 0.765625 },
  { x: 0.109375, y: 0.765625 },
  { x: 0.140625, y: 0.765625 },
  { x: 0.140625, y: 0.765625 },
  { x: 0.171875, y: 0.765625 },
  { x: 0.171875, y: 0.765625 },
  { x: 0.203125, y: 0.765625 },
  { x: 0.203125, y: 0.765625 },
  { x: 0.234375, y: 0.765625 },
  { x: 0.234375, y: 0.765625 },
  { x: 0.265625, y: 0.765625 },
  { x: 0.265625, y: 0.765625 },
  { x: 0.296875, y: 0.765625 },
  { x: 0.296875, y: 0.765625 },
  { x: 0.328125, y: 0.765625 },
  { x: 0.328125, y: 0.765625 },
  { x: 0.359375, y: 0.765625 },
  { x: 0.359375, y: 0.765625 },
  { x: 0.390625, y: 0.765625 },
  { x: 0.390625, y: 0.765625 },
  { x: 0.421875, y: 0.765625 },
  { x: 0.421875, y: 0.765625 },
  { x: 0.453125, y: 0.765625 },
  { x: 0.453125, y: 0.765625 },
  { x: 0.484375, y: 0.765625 },
  { x: 0.484375, y: 0.765625 },
  { x: 0.515625, y: 0.765625 },
  { x: 0.515625, y: 0.765625 },
  { x: 0.546875, y: 0.765625 },
  { x: 0.546875, y: 0.765625 },
  { x: 0.578125, y: 0.765625 },
  { x: 0.578125, y: 0.765625 },
  { x: 0.609375, y: 0.765625 },
  { x: 0.609375, y: 0.765625 },
  { x: 0.640625, y: 0.765625 },
  { x: 0.640625, y: 0.765625 },
  { x: 0.671875, y: 0.765625 },
  { x: 0.671875, y: 0.765625 },
  { x: 0.703125, y: 0.765625 },
  { x: 0.703125, y: 0.765625 },
  { x: 0.734375, y: 0.765625 },
  { x: 0.734375, y: 0.765625 },
  { x: 0.765625, y: 0.765625 },
  { x: 0.765625, y: 0.765625 },
  { x: 0.796875, y: 0.765625 },
  { x: 0.796875, y: 0.765625 },
  { x: 0.828125, y: 0.765625 },
  { x: 0.828125, y: 0.765625 },
  { x: 0.859375, y: 0.765625 },
  { x: 0.859375, y: 0.765625 },
  { x: 0.890625, y: 0.765625 },
  { x: 0.890625, y: 0.765625 },
  { x: 0.921875, y: 0.765625 },
  { x: 0.921875, y: 0.765625 },
  { x: 0.953125, y: 0.765625 },
  { x: 0.953125, y: 0.765625 },
  { x: 0.984375, y: 0.765625 },
  { x: 0.984375, y: 0.765625 },
  { x: 0.015625, y: 0.796875 },
  { x: 0.015625, y: 0.796875 },
  { x: 0.046875, y: 0.796875 },
  { x: 0.046875, y: 0.796875 },
  { x: 0.078125, y: 0.796875 },
  { x: 0.078125, y: 0.796875 },
  { x: 0.109375, y: 0.796875 },
  { x: 0.109375, y: 0.796875 },
  { x: 0.140625, y: 0.796875 },
  { x: 0.140625, y: 0.796875 },
  { x: 0.171875, y: 0.796875 },
  { x: 0.171875, y: 0.796875 },
  { x: 0.203125, y: 0.796875 },
  { x: 0.203125, y: 0.796875 },
  { x: 0.234375, y: 0.796875 },
  { x: 0.234375, y: 0.796875 },
  { x: 0.265625, y: 0.796875 },
  { x: 0.265625, y: 0.796875 },
  { x: 0.296875, y: 0.796875 },
  { x: 0.296875, y: 0.796875 },
  { x: 0.328125, y: 0.796875 },
  { x: 0.328125, y: 0.796875 },
  { x: 0.359375, y: 0.796875 },
  { x: 0.359375, y: 0.796875 },
  { x: 0.390625, y: 0.796875 },
  { x: 0.390625, y: 0.796875 },
  { x: 0.421875, y: 0.796875 },
  { x: 0.421875, y: 0.796875 },
  { x: 0.453125, y: 0.796875 },
  { x: 0.453125, y: 0.796875 },
  { x: 0.484375, y: 0.796875 },
  { x: 0.484375, y: 0.796875 },
  { x: 0.515625, y: 0.796875 },
  { x: 0.515625, y: 0.796875 },
  { x: 0.546875, y: 0.796875 },
  { x: 0.546875, y: 0.796875 },
  { x: 0.578125, y: 0.796875 },
  { x: 0.578125, y: 0.796875 },
  { x: 0.609375, y: 0.796875 },
  { x: 0.609375, y: 0.796875 },
  { x: 0.640625, y: 0.796875 },
  { x: 0.640625, y: 0.796875 },
  { x: 0.671875, y: 0.796875 },
  { x: 0.671875, y: 0.796875 },
  { x: 0.703125, y: 0.796875 },
  { x: 0.703125, y: 0.796875 },
  { x: 0.734375, y: 0.796875 },
  { x: 0.734375, y: 0.796875 },
  { x: 0.765625, y: 0.796875 },
  { x: 0.765625, y: 0.796875 },
  { x: 0.796875, y: 0.796875 },
  { x: 0.796875, y: 0.796875 },
  { x: 0.828125, y: 0.796875 },
  { x: 0.828125, y: 0.796875 },
  { x: 0.859375, y: 0.796875 },
  { x: 0.859375, y: 0.796875 },
  { x: 0.890625, y: 0.796875 },
  { x: 0.890625, y: 0.796875 },
  { x: 0.921875, y: 0.796875 },
  { x: 0.921875, y: 0.796875 },
  { x: 0.953125, y: 0.796875 },
  { x: 0.953125, y: 0.796875 },
  { x: 0.984375, y: 0.796875 },
  { x: 0.984375, y: 0.796875 },
  { x: 0.015625, y: 0.828125 },
  { x: 0.015625, y: 0.828125 },
  { x: 0.046875, y: 0.828125 },
  { x: 0.046875, y: 0.828125 },
  { x: 0.078125, y: 0.828125 },
  { x: 0.078125, y: 0.828125 },
  { x: 0.109375, y: 0.828125 },
  { x: 0.109375, y: 0.828125 },
  { x: 0.140625, y: 0.828125 },
  { x: 0.140625, y: 0.828125 },
  { x: 0.171875, y: 0.828125 },
  { x: 0.171875, y: 0.828125 },
  { x: 0.203125, y: 0.828125 },
  { x: 0.203125, y: 0.828125 },
  { x: 0.234375, y: 0.828125 },
  { x: 0.234375, y: 0.828125 },
  { x: 0.265625, y: 0.828125 },
  { x: 0.265625, y: 0.828125 },
  { x: 0.296875, y: 0.828125 },
  { x: 0.296875, y: 0.828125 },
  { x: 0.328125, y: 0.828125 },
  { x: 0.328125, y: 0.828125 },
  { x: 0.359375, y: 0.828125 },
  { x: 0.359375, y: 0.828125 },
  { x: 0.390625, y: 0.828125 },
  { x: 0.390625, y: 0.828125 },
  { x: 0.421875, y: 0.828125 },
  { x: 0.421875, y: 0.828125 },
  { x: 0.453125, y: 0.828125 },
  { x: 0.453125, y: 0.828125 },
  { x: 0.484375, y: 0.828125 },
  { x: 0.484375, y: 0.828125 },
  { x: 0.515625, y: 0.828125 },
  { x: 0.515625, y: 0.828125 },
  { x: 0.546875, y: 0.828125 },
  { x: 0.546875, y: 0.828125 },
  { x: 0.578125, y: 0.828125 },
  { x: 0.578125, y: 0.828125 },
  { x: 0.609375, y: 0.828125 },
  { x: 0.609375, y: 0.828125 },
  { x: 0.640625, y: 0.828125 },
  { x: 0.640625, y: 0.828125 },
  { x: 0.671875, y: 0.828125 },
  { x: 0.671875, y: 0.828125 },
  { x: 0.703125, y: 0.828125 },
  { x: 0.703125, y: 0.828125 },
  { x: 0.734375, y: 0.828125 },
  { x: 0.734375, y: 0.828125 },
  { x: 0.765625, y: 0.828125 },
  { x: 0.765625, y: 0.828125 },
  { x: 0.796875, y: 0.828125 },
  { x: 0.796875, y: 0.828125 },
  { x: 0.828125, y: 0.828125 },
  { x: 0.828125, y: 0.828125 },
  { x: 0.859375, y: 0.828125 },
  { x: 0.859375, y: 0.828125 },
  { x: 0.890625, y: 0.828125 },
  { x: 0.890625, y: 0.828125 },
  { x: 0.921875, y: 0.828125 },
  { x: 0.921875, y: 0.828125 },
  { x: 0.953125, y: 0.828125 },
  { x: 0.953125, y: 0.828125 },
  { x: 0.984375, y: 0.828125 },
  { x: 0.984375, y: 0.828125 },
  { x: 0.015625, y: 0.859375 },
  { x: 0.015625, y: 0.859375 },
  { x: 0.046875, y: 0.859375 },
  { x: 0.046875, y: 0.859375 },
  { x: 0.078125, y: 0.859375 },
  { x: 0.078125, y: 0.859375 },
  { x: 0.109375, y: 0.859375 },
  { x: 0.109375, y: 0.859375 },
  { x: 0.140625, y: 0.859375 },
  { x: 0.140625, y: 0.859375 },
  { x: 0.171875, y: 0.859375 },
  { x: 0.171875, y: 0.859375 },
  { x: 0.203125, y: 0.859375 },
  { x: 0.203125, y: 0.859375 },
  { x: 0.234375, y: 0.859375 },
  { x: 0.234375, y: 0.859375 },
  { x: 0.265625, y: 0.859375 },
  { x: 0.265625, y: 0.859375 },
  { x: 0.296875, y: 0.859375 },
  { x: 0.296875, y: 0.859375 },
  { x: 0.328125, y: 0.859375 },
  { x: 0.328125, y: 0.859375 },
  { x: 0.359375, y: 0.859375 },
  { x: 0.359375, y: 0.859375 },
  { x: 0.390625, y: 0.859375 },
  { x: 0.390625, y: 0.859375 },
  { x: 0.421875, y: 0.859375 },
  { x: 0.421875, y: 0.859375 },
  { x: 0.453125, y: 0.859375 },
  { x: 0.453125, y: 0.859375 },
  { x: 0.484375, y: 0.859375 },
  { x: 0.484375, y: 0.859375 },
  { x: 0.515625, y: 0.859375 },
  { x: 0.515625, y: 0.859375 },
  { x: 0.546875, y: 0.859375 },
  { x: 0.546875, y: 0.859375 },
  { x: 0.578125, y: 0.859375 },
  { x: 0.578125, y: 0.859375 },
  { x: 0.609375, y: 0.859375 },
  { x: 0.609375, y: 0.859375 },
  { x: 0.640625, y: 0.859375 },
  { x: 0.640625, y: 0.859375 },
  { x: 0.671875, y: 0.859375 },
  { x: 0.671875, y: 0.859375 },
  { x: 0.703125, y: 0.859375 },
  { x: 0.703125, y: 0.859375 },
  { x: 0.734375, y: 0.859375 },
  { x: 0.734375, y: 0.859375 },
  { x: 0.765625, y: 0.859375 },
  { x: 0.765625, y: 0.859375 },
  { x: 0.796875, y: 0.859375 },
  { x: 0.796875, y: 0.859375 },
  { x: 0.828125, y: 0.859375 },
  { x: 0.828125, y: 0.859375 },
  { x: 0.859375, y: 0.859375 },
  { x: 0.859375, y: 0.859375 },
  { x: 0.890625, y: 0.859375 },
  { x: 0.890625, y: 0.859375 },
  { x: 0.921875, y: 0.859375 },
  { x: 0.921875, y: 0.859375 },
  { x: 0.953125, y: 0.859375 },
  { x: 0.953125, y: 0.859375 },
  { x: 0.984375, y: 0.859375 },
  { x: 0.984375, y: 0.859375 },
  { x: 0.015625, y: 0.890625 },
  { x: 0.015625, y: 0.890625 },
  { x: 0.046875, y: 0.890625 },
  { x: 0.046875, y: 0.890625 },
  { x: 0.078125, y: 0.890625 },
  { x: 0.078125, y: 0.890625 },
  { x: 0.109375, y: 0.890625 },
  { x: 0.109375, y: 0.890625 },
  { x: 0.140625, y: 0.890625 },
  { x: 0.140625, y: 0.890625 },
  { x: 0.171875, y: 0.890625 },
  { x: 0.171875, y: 0.890625 },
  { x: 0.203125, y: 0.890625 },
  { x: 0.203125, y: 0.890625 },
  { x: 0.234375, y: 0.890625 },
  { x: 0.234375, y: 0.890625 },
  { x: 0.265625, y: 0.890625 },
  { x: 0.265625, y: 0.890625 },
  { x: 0.296875, y: 0.890625 },
  { x: 0.296875, y: 0.890625 },
  { x: 0.328125, y: 0.890625 },
  { x: 0.328125, y: 0.890625 },
  { x: 0.359375, y: 0.890625 },
  { x: 0.359375, y: 0.890625 },
  { x: 0.390625, y: 0.890625 },
  { x: 0.390625, y: 0.890625 },
  { x: 0.421875, y: 0.890625 },
  { x: 0.421875, y: 0.890625 },
  { x: 0.453125, y: 0.890625 },
  { x: 0.453125, y: 0.890625 },
  { x: 0.484375, y: 0.890625 },
  { x: 0.484375, y: 0.890625 },
  { x: 0.515625, y: 0.890625 },
  { x: 0.515625, y: 0.890625 },
  { x: 0.546875, y: 0.890625 },
  { x: 0.546875, y: 0.890625 },
  { x: 0.578125, y: 0.890625 },
  { x: 0.578125, y: 0.890625 },
  { x: 0.609375, y: 0.890625 },
  { x: 0.609375, y: 0.890625 },
  { x: 0.640625, y: 0.890625 },
  { x: 0.640625, y: 0.890625 },
  { x: 0.671875, y: 0.890625 },
  { x: 0.671875, y: 0.890625 },
  { x: 0.703125, y: 0.890625 },
  { x: 0.703125, y: 0.890625 },
  { x: 0.734375, y: 0.890625 },
  { x: 0.734375, y: 0.890625 },
  { x: 0.765625, y: 0.890625 },
  { x: 0.765625, y: 0.890625 },
  { x: 0.796875, y: 0.890625 },
  { x: 0.796875, y: 0.890625 },
  { x: 0.828125, y: 0.890625 },
  { x: 0.828125, y: 0.890625 },
  { x: 0.859375, y: 0.890625 },
  { x: 0.859375, y: 0.890625 },
  { x: 0.890625, y: 0.890625 },
  { x: 0.890625, y: 0.890625 },
  { x: 0.921875, y: 0.890625 },
  { x: 0.921875, y: 0.890625 },
  { x: 0.953125, y: 0.890625 },
  { x: 0.953125, y: 0.890625 },
  { x: 0.984375, y: 0.890625 },
  { x: 0.984375, y: 0.890625 },
  { x: 0.015625, y: 0.921875 },
  { x: 0.015625, y: 0.921875 },
  { x: 0.046875, y: 0.921875 },
  { x: 0.046875, y: 0.921875 },
  { x: 0.078125, y: 0.921875 },
  { x: 0.078125, y: 0.921875 },
  { x: 0.109375, y: 0.921875 },
  { x: 0.109375, y: 0.921875 },
  { x: 0.140625, y: 0.921875 },
  { x: 0.140625, y: 0.921875 },
  { x: 0.171875, y: 0.921875 },
  { x: 0.171875, y: 0.921875 },
  { x: 0.203125, y: 0.921875 },
  { x: 0.203125, y: 0.921875 },
  { x: 0.234375, y: 0.921875 },
  { x: 0.234375, y: 0.921875 },
  { x: 0.265625, y: 0.921875 },
  { x: 0.265625, y: 0.921875 },
  { x: 0.296875, y: 0.921875 },
  { x: 0.296875, y: 0.921875 },
  { x: 0.328125, y: 0.921875 },
  { x: 0.328125, y: 0.921875 },
  { x: 0.359375, y: 0.921875 },
  { x: 0.359375, y: 0.921875 },
  { x: 0.390625, y: 0.921875 },
  { x: 0.390625, y: 0.921875 },
  { x: 0.421875, y: 0.921875 },
  { x: 0.421875, y: 0.921875 },
  { x: 0.453125, y: 0.921875 },
  { x: 0.453125, y: 0.921875 },
  { x: 0.484375, y: 0.921875 },
  { x: 0.484375, y: 0.921875 },
  { x: 0.515625, y: 0.921875 },
  { x: 0.515625, y: 0.921875 },
  { x: 0.546875, y: 0.921875 },
  { x: 0.546875, y: 0.921875 },
  { x: 0.578125, y: 0.921875 },
  { x: 0.578125, y: 0.921875 },
  { x: 0.609375, y: 0.921875 },
  { x: 0.609375, y: 0.921875 },
  { x: 0.640625, y: 0.921875 },
  { x: 0.640625, y: 0.921875 },
  { x: 0.671875, y: 0.921875 },
  { x: 0.671875, y: 0.921875 },
  { x: 0.703125, y: 0.921875 },
  { x: 0.703125, y: 0.921875 },
  { x: 0.734375, y: 0.921875 },
  { x: 0.734375, y: 0.921875 },
  { x: 0.765625, y: 0.921875 },
  { x: 0.765625, y: 0.921875 },
  { x: 0.796875, y: 0.921875 },
  { x: 0.796875, y: 0.921875 },
  { x: 0.828125, y: 0.921875 },
  { x: 0.828125, y: 0.921875 },
  { x: 0.859375, y: 0.921875 },
  { x: 0.859375, y: 0.921875 },
  { x: 0.890625, y: 0.921875 },
  { x: 0.890625, y: 0.921875 },
  { x: 0.921875, y: 0.921875 },
  { x: 0.921875, y: 0.921875 },
  { x: 0.953125, y: 0.921875 },
  { x: 0.953125, y: 0.921875 },
  { x: 0.984375, y: 0.921875 },
  { x: 0.984375, y: 0.921875 },
  { x: 0.015625, y: 0.953125 },
  { x: 0.015625, y: 0.953125 },
  { x: 0.046875, y: 0.953125 },
  { x: 0.046875, y: 0.953125 },
  { x: 0.078125, y: 0.953125 },
  { x: 0.078125, y: 0.953125 },
  { x: 0.109375, y: 0.953125 },
  { x: 0.109375, y: 0.953125 },
  { x: 0.140625, y: 0.953125 },
  { x: 0.140625, y: 0.953125 },
  { x: 0.171875, y: 0.953125 },
  { x: 0.171875, y: 0.953125 },
  { x: 0.203125, y: 0.953125 },
  { x: 0.203125, y: 0.953125 },
  { x: 0.234375, y: 0.953125 },
  { x: 0.234375, y: 0.953125 },
  { x: 0.265625, y: 0.953125 },
  { x: 0.265625, y: 0.953125 },
  { x: 0.296875, y: 0.953125 },
  { x: 0.296875, y: 0.953125 },
  { x: 0.328125, y: 0.953125 },
  { x: 0.328125, y: 0.953125 },
  { x: 0.359375, y: 0.953125 },
  { x: 0.359375, y: 0.953125 },
  { x: 0.390625, y: 0.953125 },
  { x: 0.390625, y: 0.953125 },
  { x: 0.421875, y: 0.953125 },
  { x: 0.421875, y: 0.953125 },
  { x: 0.453125, y: 0.953125 },
  { x: 0.453125, y: 0.953125 },
  { x: 0.484375, y: 0.953125 },
  { x: 0.484375, y: 0.953125 },
  { x: 0.515625, y: 0.953125 },
  { x: 0.515625, y: 0.953125 },
  { x: 0.546875, y: 0.953125 },
  { x: 0.546875, y: 0.953125 },
  { x: 0.578125, y: 0.953125 },
  { x: 0.578125, y: 0.953125 },
  { x: 0.609375, y: 0.953125 },
  { x: 0.609375, y: 0.953125 },
  { x: 0.640625, y: 0.953125 },
  { x: 0.640625, y: 0.953125 },
  { x: 0.671875, y: 0.953125 },
  { x: 0.671875, y: 0.953125 },
  { x: 0.703125, y: 0.953125 },
  { x: 0.703125, y: 0.953125 },
  { x: 0.734375, y: 0.953125 },
  { x: 0.734375, y: 0.953125 },
  { x: 0.765625, y: 0.953125 },
  { x: 0.765625, y: 0.953125 },
  { x: 0.796875, y: 0.953125 },
  { x: 0.796875, y: 0.953125 },
  { x: 0.828125, y: 0.953125 },
  { x: 0.828125, y: 0.953125 },
  { x: 0.859375, y: 0.953125 },
  { x: 0.859375, y: 0.953125 },
  { x: 0.890625, y: 0.953125 },
  { x: 0.890625, y: 0.953125 },
  { x: 0.921875, y: 0.953125 },
  { x: 0.921875, y: 0.953125 },
  { x: 0.953125, y: 0.953125 },
  { x: 0.953125, y: 0.953125 },
  { x: 0.984375, y: 0.953125 },
  { x: 0.984375, y: 0.953125 },
  { x: 0.015625, y: 0.984375 },
  { x: 0.015625, y: 0.984375 },
  { x: 0.046875, y: 0.984375 },
  { x: 0.046875, y: 0.984375 },
  { x: 0.078125, y: 0.984375 },
  { x: 0.078125, y: 0.984375 },
  { x: 0.109375, y: 0.984375 },
  { x: 0.109375, y: 0.984375 },
  { x: 0.140625, y: 0.984375 },
  { x: 0.140625, y: 0.984375 },
  { x: 0.171875, y: 0.984375 },
  { x: 0.171875, y: 0.984375 },
  { x: 0.203125, y: 0.984375 },
  { x: 0.203125, y: 0.984375 },
  { x: 0.234375, y: 0.984375 },
  { x: 0.234375, y: 0.984375 },
  { x: 0.265625, y: 0.984375 },
  { x: 0.265625, y: 0.984375 },
  { x: 0.296875, y: 0.984375 },
  { x: 0.296875, y: 0.984375 },
  { x: 0.328125, y: 0.984375 },
  { x: 0.328125, y: 0.984375 },
  { x: 0.359375, y: 0.984375 },
  { x: 0.359375, y: 0.984375 },
  { x: 0.390625, y: 0.984375 },
  { x: 0.390625, y: 0.984375 },
  { x: 0.421875, y: 0.984375 },
  { x: 0.421875, y: 0.984375 },
  { x: 0.453125, y: 0.984375 },
  { x: 0.453125, y: 0.984375 },
  { x: 0.484375, y: 0.984375 },
  { x: 0.484375, y: 0.984375 },
  { x: 0.515625, y: 0.984375 },
  { x: 0.515625, y: 0.984375 },
  { x: 0.546875, y: 0.984375 },
  { x: 0.546875, y: 0.984375 },
  { x: 0.578125, y: 0.984375 },
  { x: 0.578125, y: 0.984375 },
  { x: 0.609375, y: 0.984375 },
  { x: 0.609375, y: 0.984375 },
  { x: 0.640625, y: 0.984375 },
  { x: 0.640625, y: 0.984375 },
  { x: 0.671875, y: 0.984375 },
  { x: 0.671875, y: 0.984375 },
  { x: 0.703125, y: 0.984375 },
  { x: 0.703125, y: 0.984375 },
  { x: 0.734375, y: 0.984375 },
  { x: 0.734375, y: 0.984375 },
  { x: 0.765625, y: 0.984375 },
  { x: 0.765625, y: 0.984375 },
  { x: 0.796875, y: 0.984375 },
  { x: 0.796875, y: 0.984375 },
  { x: 0.828125, y: 0.984375 },
  { x: 0.828125, y: 0.984375 },
  { x: 0.859375, y: 0.984375 },
  { x: 0.859375, y: 0.984375 },
  { x: 0.890625, y: 0.984375 },
  { x: 0.890625, y: 0.984375 },
  { x: 0.921875, y: 0.984375 },
  { x: 0.921875, y: 0.984375 },
  { x: 0.953125, y: 0.984375 },
  { x: 0.953125, y: 0.984375 },
  { x: 0.984375, y: 0.984375 },
  { x: 0.984375, y: 0.984375 },
  { x: 0.03125, y: 0.03125 },
  { x: 0.03125, y: 0.03125 },
  { x: 0.09375, y: 0.03125 },
  { x: 0.09375, y: 0.03125 },
  { x: 0.15625, y: 0.03125 },
  { x: 0.15625, y: 0.03125 },
  { x: 0.21875, y: 0.03125 },
  { x: 0.21875, y: 0.03125 },
  { x: 0.28125, y: 0.03125 },
  { x: 0.28125, y: 0.03125 },
  { x: 0.34375, y: 0.03125 },
  { x: 0.34375, y: 0.03125 },
  { x: 0.40625, y: 0.03125 },
  { x: 0.40625, y: 0.03125 },
  { x: 0.46875, y: 0.03125 },
  { x: 0.46875, y: 0.03125 },
  { x: 0.53125, y: 0.03125 },
  { x: 0.53125, y: 0.03125 },
  { x: 0.59375, y: 0.03125 },
  { x: 0.59375, y: 0.03125 },
  { x: 0.65625, y: 0.03125 },
  { x: 0.65625, y: 0.03125 },
  { x: 0.71875, y: 0.03125 },
  { x: 0.71875, y: 0.03125 },
  { x: 0.78125, y: 0.03125 },
  { x: 0.78125, y: 0.03125 },
  { x: 0.84375, y: 0.03125 },
  { x: 0.84375, y: 0.03125 },
  { x: 0.90625, y: 0.03125 },
  { x: 0.90625, y: 0.03125 },
  { x: 0.96875, y: 0.03125 },
  { x: 0.96875, y: 0.03125 },
  { x: 0.03125, y: 0.09375 },
  { x: 0.03125, y: 0.09375 },
  { x: 0.09375, y: 0.09375 },
  { x: 0.09375, y: 0.09375 },
  { x: 0.15625, y: 0.09375 },
  { x: 0.15625, y: 0.09375 },
  { x: 0.21875, y: 0.09375 },
  { x: 0.21875, y: 0.09375 },
  { x: 0.28125, y: 0.09375 },
  { x: 0.28125, y: 0.09375 },
  { x: 0.34375, y: 0.09375 },
  { x: 0.34375, y: 0.09375 },
  { x: 0.40625, y: 0.09375 },
  { x: 0.40625, y: 0.09375 },
  { x: 0.46875, y: 0.09375 },
  { x: 0.46875, y: 0.09375 },
  { x: 0.53125, y: 0.09375 },
  { x: 0.53125, y: 0.09375 },
  { x: 0.59375, y: 0.09375 },
  { x: 0.59375, y: 0.09375 },
  { x: 0.65625, y: 0.09375 },
  { x: 0.65625, y: 0.09375 },
  { x: 0.71875, y: 0.09375 },
  { x: 0.71875, y: 0.09375 },
  { x: 0.78125, y: 0.09375 },
  { x: 0.78125, y: 0.09375 },
  { x: 0.84375, y: 0.09375 },
  { x: 0.84375, y: 0.09375 },
  { x: 0.90625, y: 0.09375 },
  { x: 0.90625, y: 0.09375 },
  { x: 0.96875, y: 0.09375 },
  { x: 0.96875, y: 0.09375 },
  { x: 0.03125, y: 0.15625 },
  { x: 0.03125, y: 0.15625 },
  { x: 0.09375, y: 0.15625 },
  { x: 0.09375, y: 0.15625 },
  { x: 0.15625, y: 0.15625 },
  { x: 0.15625, y: 0.15625 },
  { x: 0.21875, y: 0.15625 },
  { x: 0.21875, y: 0.15625 },
  { x: 0.28125, y: 0.15625 },
  { x: 0.28125, y: 0.15625 },
  { x: 0.34375, y: 0.15625 },
  { x: 0.34375, y: 0.15625 },
  { x: 0.40625, y: 0.15625 },
  { x: 0.40625, y: 0.15625 },
  { x: 0.46875, y: 0.15625 },
  { x: 0.46875, y: 0.15625 },
  { x: 0.53125, y: 0.15625 },
  { x: 0.53125, y: 0.15625 },
  { x: 0.59375, y: 0.15625 },
  { x: 0.59375, y: 0.15625 },
  { x: 0.65625, y: 0.15625 },
  { x: 0.65625, y: 0.15625 },
  { x: 0.71875, y: 0.15625 },
  { x: 0.71875, y: 0.15625 },
  { x: 0.78125, y: 0.15625 },
  { x: 0.78125, y: 0.15625 },
  { x: 0.84375, y: 0.15625 },
  { x: 0.84375, y: 0.15625 },
  { x: 0.90625, y: 0.15625 },
  { x: 0.90625, y: 0.15625 },
  { x: 0.96875, y: 0.15625 },
  { x: 0.96875, y: 0.15625 },
  { x: 0.03125, y: 0.21875 },
  { x: 0.03125, y: 0.21875 },
  { x: 0.09375, y: 0.21875 },
  { x: 0.09375, y: 0.21875 },
  { x: 0.15625, y: 0.21875 },
  { x: 0.15625, y: 0.21875 },
  { x: 0.21875, y: 0.21875 },
  { x: 0.21875, y: 0.21875 },
  { x: 0.28125, y: 0.21875 },
  { x: 0.28125, y: 0.21875 },
  { x: 0.34375, y: 0.21875 },
  { x: 0.34375, y: 0.21875 },
  { x: 0.40625, y: 0.21875 },
  { x: 0.40625, y: 0.21875 },
  { x: 0.46875, y: 0.21875 },
  { x: 0.46875, y: 0.21875 },
  { x: 0.53125, y: 0.21875 },
  { x: 0.53125, y: 0.21875 },
  { x: 0.59375, y: 0.21875 },
  { x: 0.59375, y: 0.21875 },
  { x: 0.65625, y: 0.21875 },
  { x: 0.65625, y: 0.21875 },
  { x: 0.71875, y: 0.21875 },
  { x: 0.71875, y: 0.21875 },
  { x: 0.78125, y: 0.21875 },
  { x: 0.78125, y: 0.21875 },
  { x: 0.84375, y: 0.21875 },
  { x: 0.84375, y: 0.21875 },
  { x: 0.90625, y: 0.21875 },
  { x: 0.90625, y: 0.21875 },
  { x: 0.96875, y: 0.21875 },
  { x: 0.96875, y: 0.21875 },
  { x: 0.03125, y: 0.28125 },
  { x: 0.03125, y: 0.28125 },
  { x: 0.09375, y: 0.28125 },
  { x: 0.09375, y: 0.28125 },
  { x: 0.15625, y: 0.28125 },
  { x: 0.15625, y: 0.28125 },
  { x: 0.21875, y: 0.28125 },
  { x: 0.21875, y: 0.28125 },
  { x: 0.28125, y: 0.28125 },
  { x: 0.28125, y: 0.28125 },
  { x: 0.34375, y: 0.28125 },
  { x: 0.34375, y: 0.28125 },
  { x: 0.40625, y: 0.28125 },
  { x: 0.40625, y: 0.28125 },
  { x: 0.46875, y: 0.28125 },
  { x: 0.46875, y: 0.28125 },
  { x: 0.53125, y: 0.28125 },
  { x: 0.53125, y: 0.28125 },
  { x: 0.59375, y: 0.28125 },
  { x: 0.59375, y: 0.28125 },
  { x: 0.65625, y: 0.28125 },
  { x: 0.65625, y: 0.28125 },
  { x: 0.71875, y: 0.28125 },
  { x: 0.71875, y: 0.28125 },
  { x: 0.78125, y: 0.28125 },
  { x: 0.78125, y: 0.28125 },
  { x: 0.84375, y: 0.28125 },
  { x: 0.84375, y: 0.28125 },
  { x: 0.90625, y: 0.28125 },
  { x: 0.90625, y: 0.28125 },
  { x: 0.96875, y: 0.28125 },
  { x: 0.96875, y: 0.28125 },
  { x: 0.03125, y: 0.34375 },
  { x: 0.03125, y: 0.34375 },
  { x: 0.09375, y: 0.34375 },
  { x: 0.09375, y: 0.34375 },
  { x: 0.15625, y: 0.34375 },
  { x: 0.15625, y: 0.34375 },
  { x: 0.21875, y: 0.34375 },
  { x: 0.21875, y: 0.34375 },
  { x: 0.28125, y: 0.34375 },
  { x: 0.28125, y: 0.34375 },
  { x: 0.34375, y: 0.34375 },
  { x: 0.34375, y: 0.34375 },
  { x: 0.40625, y: 0.34375 },
  { x: 0.40625, y: 0.34375 },
  { x: 0.46875, y: 0.34375 },
  { x: 0.46875, y: 0.34375 },
  { x: 0.53125, y: 0.34375 },
  { x: 0.53125, y: 0.34375 },
  { x: 0.59375, y: 0.34375 },
  { x: 0.59375, y: 0.34375 },
  { x: 0.65625, y: 0.34375 },
  { x: 0.65625, y: 0.34375 },
  { x: 0.71875, y: 0.34375 },
  { x: 0.71875, y: 0.34375 },
  { x: 0.78125, y: 0.34375 },
  { x: 0.78125, y: 0.34375 },
  { x: 0.84375, y: 0.34375 },
  { x: 0.84375, y: 0.34375 },
  { x: 0.90625, y: 0.34375 },
  { x: 0.90625, y: 0.34375 },
  { x: 0.96875, y: 0.34375 },
  { x: 0.96875, y: 0.34375 },
  { x: 0.03125, y: 0.40625 },
  { x: 0.03125, y: 0.40625 },
  { x: 0.09375, y: 0.40625 },
  { x: 0.09375, y: 0.40625 },
  { x: 0.15625, y: 0.40625 },
  { x: 0.15625, y: 0.40625 },
  { x: 0.21875, y: 0.40625 },
  { x: 0.21875, y: 0.40625 },
  { x: 0.28125, y: 0.40625 },
  { x: 0.28125, y: 0.40625 },
  { x: 0.34375, y: 0.40625 },
  { x: 0.34375, y: 0.40625 },
  { x: 0.40625, y: 0.40625 },
  { x: 0.40625, y: 0.40625 },
  { x: 0.46875, y: 0.40625 },
  { x: 0.46875, y: 0.40625 },
  { x: 0.53125, y: 0.40625 },
  { x: 0.53125, y: 0.40625 },
  { x: 0.59375, y: 0.40625 },
  { x: 0.59375, y: 0.40625 },
  { x: 0.65625, y: 0.40625 },
  { x: 0.65625, y: 0.40625 },
  { x: 0.71875, y: 0.40625 },
  { x: 0.71875, y: 0.40625 },
  { x: 0.78125, y: 0.40625 },
  { x: 0.78125, y: 0.40625 },
  { x: 0.84375, y: 0.40625 },
  { x: 0.84375, y: 0.40625 },
  { x: 0.90625, y: 0.40625 },
  { x: 0.90625, y: 0.40625 },
  { x: 0.96875, y: 0.40625 },
  { x: 0.96875, y: 0.40625 },
  { x: 0.03125, y: 0.46875 },
  { x: 0.03125, y: 0.46875 },
  { x: 0.09375, y: 0.46875 },
  { x: 0.09375, y: 0.46875 },
  { x: 0.15625, y: 0.46875 },
  { x: 0.15625, y: 0.46875 },
  { x: 0.21875, y: 0.46875 },
  { x: 0.21875, y: 0.46875 },
  { x: 0.28125, y: 0.46875 },
  { x: 0.28125, y: 0.46875 },
  { x: 0.34375, y: 0.46875 },
  { x: 0.34375, y: 0.46875 },
  { x: 0.40625, y: 0.46875 },
  { x: 0.40625, y: 0.46875 },
  { x: 0.46875, y: 0.46875 },
  { x: 0.46875, y: 0.46875 },
  { x: 0.53125, y: 0.46875 },
  { x: 0.53125, y: 0.46875 },
  { x: 0.59375, y: 0.46875 },
  { x: 0.59375, y: 0.46875 },
  { x: 0.65625, y: 0.46875 },
  { x: 0.65625, y: 0.46875 },
  { x: 0.71875, y: 0.46875 },
  { x: 0.71875, y: 0.46875 },
  { x: 0.78125, y: 0.46875 },
  { x: 0.78125, y: 0.46875 },
  { x: 0.84375, y: 0.46875 },
  { x: 0.84375, y: 0.46875 },
  { x: 0.90625, y: 0.46875 },
  { x: 0.90625, y: 0.46875 },
  { x: 0.96875, y: 0.46875 },
  { x: 0.96875, y: 0.46875 },
  { x: 0.03125, y: 0.53125 },
  { x: 0.03125, y: 0.53125 },
  { x: 0.09375, y: 0.53125 },
  { x: 0.09375, y: 0.53125 },
  { x: 0.15625, y: 0.53125 },
  { x: 0.15625, y: 0.53125 },
  { x: 0.21875, y: 0.53125 },
  { x: 0.21875, y: 0.53125 },
  { x: 0.28125, y: 0.53125 },
  { x: 0.28125, y: 0.53125 },
  { x: 0.34375, y: 0.53125 },
  { x: 0.34375, y: 0.53125 },
  { x: 0.40625, y: 0.53125 },
  { x: 0.40625, y: 0.53125 },
  { x: 0.46875, y: 0.53125 },
  { x: 0.46875, y: 0.53125 },
  { x: 0.53125, y: 0.53125 },
  { x: 0.53125, y: 0.53125 },
  { x: 0.59375, y: 0.53125 },
  { x: 0.59375, y: 0.53125 },
  { x: 0.65625, y: 0.53125 },
  { x: 0.65625, y: 0.53125 },
  { x: 0.71875, y: 0.53125 },
  { x: 0.71875, y: 0.53125 },
  { x: 0.78125, y: 0.53125 },
  { x: 0.78125, y: 0.53125 },
  { x: 0.84375, y: 0.53125 },
  { x: 0.84375, y: 0.53125 },
  { x: 0.90625, y: 0.53125 },
  { x: 0.90625, y: 0.53125 },
  { x: 0.96875, y: 0.53125 },
  { x: 0.96875, y: 0.53125 },
  { x: 0.03125, y: 0.59375 },
  { x: 0.03125, y: 0.59375 },
  { x: 0.09375, y: 0.59375 },
  { x: 0.09375, y: 0.59375 },
  { x: 0.15625, y: 0.59375 },
  { x: 0.15625, y: 0.59375 },
  { x: 0.21875, y: 0.59375 },
  { x: 0.21875, y: 0.59375 },
  { x: 0.28125, y: 0.59375 },
  { x: 0.28125, y: 0.59375 },
  { x: 0.34375, y: 0.59375 },
  { x: 0.34375, y: 0.59375 },
  { x: 0.40625, y: 0.59375 },
  { x: 0.40625, y: 0.59375 },
  { x: 0.46875, y: 0.59375 },
  { x: 0.46875, y: 0.59375 },
  { x: 0.53125, y: 0.59375 },
  { x: 0.53125, y: 0.59375 },
  { x: 0.59375, y: 0.59375 },
  { x: 0.59375, y: 0.59375 },
  { x: 0.65625, y: 0.59375 },
  { x: 0.65625, y: 0.59375 },
  { x: 0.71875, y: 0.59375 },
  { x: 0.71875, y: 0.59375 },
  { x: 0.78125, y: 0.59375 },
  { x: 0.78125, y: 0.59375 },
  { x: 0.84375, y: 0.59375 },
  { x: 0.84375, y: 0.59375 },
  { x: 0.90625, y: 0.59375 },
  { x: 0.90625, y: 0.59375 },
  { x: 0.96875, y: 0.59375 },
  { x: 0.96875, y: 0.59375 },
  { x: 0.03125, y: 0.65625 },
  { x: 0.03125, y: 0.65625 },
  { x: 0.09375, y: 0.65625 },
  { x: 0.09375, y: 0.65625 },
  { x: 0.15625, y: 0.65625 },
  { x: 0.15625, y: 0.65625 },
  { x: 0.21875, y: 0.65625 },
  { x: 0.21875, y: 0.65625 },
  { x: 0.28125, y: 0.65625 },
  { x: 0.28125, y: 0.65625 },
  { x: 0.34375, y: 0.65625 },
  { x: 0.34375, y: 0.65625 },
  { x: 0.40625, y: 0.65625 },
  { x: 0.40625, y: 0.65625 },
  { x: 0.46875, y: 0.65625 },
  { x: 0.46875, y: 0.65625 },
  { x: 0.53125, y: 0.65625 },
  { x: 0.53125, y: 0.65625 },
  { x: 0.59375, y: 0.65625 },
  { x: 0.59375, y: 0.65625 },
  { x: 0.65625, y: 0.65625 },
  { x: 0.65625, y: 0.65625 },
  { x: 0.71875, y: 0.65625 },
  { x: 0.71875, y: 0.65625 },
  { x: 0.78125, y: 0.65625 },
  { x: 0.78125, y: 0.65625 },
  { x: 0.84375, y: 0.65625 },
  { x: 0.84375, y: 0.65625 },
  { x: 0.90625, y: 0.65625 },
  { x: 0.90625, y: 0.65625 },
  { x: 0.96875, y: 0.65625 },
  { x: 0.96875, y: 0.65625 },
  { x: 0.03125, y: 0.71875 },
  { x: 0.03125, y: 0.71875 },
  { x: 0.09375, y: 0.71875 },
  { x: 0.09375, y: 0.71875 },
  { x: 0.15625, y: 0.71875 },
  { x: 0.15625, y: 0.71875 },
  { x: 0.21875, y: 0.71875 },
  { x: 0.21875, y: 0.71875 },
  { x: 0.28125, y: 0.71875 },
  { x: 0.28125, y: 0.71875 },
  { x: 0.34375, y: 0.71875 },
  { x: 0.34375, y: 0.71875 },
  { x: 0.40625, y: 0.71875 },
  { x: 0.40625, y: 0.71875 },
  { x: 0.46875, y: 0.71875 },
  { x: 0.46875, y: 0.71875 },
  { x: 0.53125, y: 0.71875 },
  { x: 0.53125, y: 0.71875 },
  { x: 0.59375, y: 0.71875 },
  { x: 0.59375, y: 0.71875 },
  { x: 0.65625, y: 0.71875 },
  { x: 0.65625, y: 0.71875 },
  { x: 0.71875, y: 0.71875 },
  { x: 0.71875, y: 0.71875 },
  { x: 0.78125, y: 0.71875 },
  { x: 0.78125, y: 0.71875 },
  { x: 0.84375, y: 0.71875 },
  { x: 0.84375, y: 0.71875 },
  { x: 0.90625, y: 0.71875 },
  { x: 0.90625, y: 0.71875 },
  { x: 0.96875, y: 0.71875 },
  { x: 0.96875, y: 0.71875 },
  { x: 0.03125, y: 0.78125 },
  { x: 0.03125, y: 0.78125 },
  { x: 0.09375, y: 0.78125 },
  { x: 0.09375, y: 0.78125 },
  { x: 0.15625, y: 0.78125 },
  { x: 0.15625, y: 0.78125 },
  { x: 0.21875, y: 0.78125 },
  { x: 0.21875, y: 0.78125 },
  { x: 0.28125, y: 0.78125 },
  { x: 0.28125, y: 0.78125 },
  { x: 0.34375, y: 0.78125 },
  { x: 0.34375, y: 0.78125 },
  { x: 0.40625, y: 0.78125 },
  { x: 0.40625, y: 0.78125 },
  { x: 0.46875, y: 0.78125 },
  { x: 0.46875, y: 0.78125 },
  { x: 0.53125, y: 0.78125 },
  { x: 0.53125, y: 0.78125 },
  { x: 0.59375, y: 0.78125 },
  { x: 0.59375, y: 0.78125 },
  { x: 0.65625, y: 0.78125 },
  { x: 0.65625, y: 0.78125 },
  { x: 0.71875, y: 0.78125 },
  { x: 0.71875, y: 0.78125 },
  { x: 0.78125, y: 0.78125 },
  { x: 0.78125, y: 0.78125 },
  { x: 0.84375, y: 0.78125 },
  { x: 0.84375, y: 0.78125 },
  { x: 0.90625, y: 0.78125 },
  { x: 0.90625, y: 0.78125 },
  { x: 0.96875, y: 0.78125 },
  { x: 0.96875, y: 0.78125 },
  { x: 0.03125, y: 0.84375 },
  { x: 0.03125, y: 0.84375 },
  { x: 0.09375, y: 0.84375 },
  { x: 0.09375, y: 0.84375 },
  { x: 0.15625, y: 0.84375 },
  { x: 0.15625, y: 0.84375 },
  { x: 0.21875, y: 0.84375 },
  { x: 0.21875, y: 0.84375 },
  { x: 0.28125, y: 0.84375 },
  { x: 0.28125, y: 0.84375 },
  { x: 0.34375, y: 0.84375 },
  { x: 0.34375, y: 0.84375 },
  { x: 0.40625, y: 0.84375 },
  { x: 0.40625, y: 0.84375 },
  { x: 0.46875, y: 0.84375 },
  { x: 0.46875, y: 0.84375 },
  { x: 0.53125, y: 0.84375 },
  { x: 0.53125, y: 0.84375 },
  { x: 0.59375, y: 0.84375 },
  { x: 0.59375, y: 0.84375 },
  { x: 0.65625, y: 0.84375 },
  { x: 0.65625, y: 0.84375 },
  { x: 0.71875, y: 0.84375 },
  { x: 0.71875, y: 0.84375 },
  { x: 0.78125, y: 0.84375 },
  { x: 0.78125, y: 0.84375 },
  { x: 0.84375, y: 0.84375 },
  { x: 0.84375, y: 0.84375 },
  { x: 0.90625, y: 0.84375 },
  { x: 0.90625, y: 0.84375 },
  { x: 0.96875, y: 0.84375 },
  { x: 0.96875, y: 0.84375 },
  { x: 0.03125, y: 0.90625 },
  { x: 0.03125, y: 0.90625 },
  { x: 0.09375, y: 0.90625 },
  { x: 0.09375, y: 0.90625 },
  { x: 0.15625, y: 0.90625 },
  { x: 0.15625, y: 0.90625 },
  { x: 0.21875, y: 0.90625 },
  { x: 0.21875, y: 0.90625 },
  { x: 0.28125, y: 0.90625 },
  { x: 0.28125, y: 0.90625 },
  { x: 0.34375, y: 0.90625 },
  { x: 0.34375, y: 0.90625 },
  { x: 0.40625, y: 0.90625 },
  { x: 0.40625, y: 0.90625 },
  { x: 0.46875, y: 0.90625 },
  { x: 0.46875, y: 0.90625 },
  { x: 0.53125, y: 0.90625 },
  { x: 0.53125, y: 0.90625 },
  { x: 0.59375, y: 0.90625 },
  { x: 0.59375, y: 0.90625 },
  { x: 0.65625, y: 0.90625 },
  { x: 0.65625, y: 0.90625 },
  { x: 0.71875, y: 0.90625 },
  { x: 0.71875, y: 0.90625 },
  { x: 0.78125, y: 0.90625 },
  { x: 0.78125, y: 0.90625 },
  { x: 0.84375, y: 0.90625 },
  { x: 0.84375, y: 0.90625 },
  { x: 0.90625, y: 0.90625 },
  { x: 0.90625, y: 0.90625 },
  { x: 0.96875, y: 0.90625 },
  { x: 0.96875, y: 0.90625 },
  { x: 0.03125, y: 0.96875 },
  { x: 0.03125, y: 0.96875 },
  { x: 0.09375, y: 0.96875 },
  { x: 0.09375, y: 0.96875 },
  { x: 0.15625, y: 0.96875 },
  { x: 0.15625, y: 0.96875 },
  { x: 0.21875, y: 0.96875 },
  { x: 0.21875, y: 0.96875 },
  { x: 0.28125, y: 0.96875 },
  { x: 0.28125, y: 0.96875 },
  { x: 0.34375, y: 0.96875 },
  { x: 0.34375, y: 0.96875 },
  { x: 0.40625, y: 0.96875 },
  { x: 0.40625, y: 0.96875 },
  { x: 0.46875, y: 0.96875 },
  { x: 0.46875, y: 0.96875 },
  { x: 0.53125, y: 0.96875 },
  { x: 0.53125, y: 0.96875 },
  { x: 0.59375, y: 0.96875 },
  { x: 0.59375, y: 0.96875 },
  { x: 0.65625, y: 0.96875 },
  { x: 0.65625, y: 0.96875 },
  { x: 0.71875, y: 0.96875 },
  { x: 0.71875, y: 0.96875 },
  { x: 0.78125, y: 0.96875 },
  { x: 0.78125, y: 0.96875 },
  { x: 0.84375, y: 0.96875 },
  { x: 0.84375, y: 0.96875 },
  { x: 0.90625, y: 0.96875 },
  { x: 0.90625, y: 0.96875 },
  { x: 0.96875, y: 0.96875 },
  { x: 0.96875, y: 0.96875 },
  { x: 0.0625, y: 0.0625 },
  { x: 0.0625, y: 0.0625 },
  { x: 0.0625, y: 0.0625 },
  { x: 0.0625, y: 0.0625 },
  { x: 0.0625, y: 0.0625 },
  { x: 0.0625, y: 0.0625 },
  { x: 0.1875, y: 0.0625 },
  { x: 0.1875, y: 0.0625 },
  { x: 0.1875, y: 0.0625 },
  { x: 0.1875, y: 0.0625 },
  { x: 0.1875, y: 0.0625 },
  { x: 0.1875, y: 0.0625 },
  { x: 0.3125, y: 0.0625 },
  { x: 0.3125, y: 0.0625 },
  { x: 0.3125, y: 0.0625 },
  { x: 0.3125, y: 0.0625 },
  { x: 0.3125, y: 0.0625 },
  { x: 0.3125, y: 0.0625 },
  { x: 0.4375, y: 0.0625 },
  { x: 0.4375, y: 0.0625 },
  { x: 0.4375, y: 0.0625 },
  { x: 0.4375, y: 0.0625 },
  { x: 0.4375, y: 0.0625 },
  { x: 0.4375, y: 0.0625 },
  { x: 0.5625, y: 0.0625 },
  { x: 0.5625, y: 0.0625 },
  { x: 0.5625, y: 0.0625 },
  { x: 0.5625, y: 0.0625 },
  { x: 0.5625, y: 0.0625 },
  { x: 0.5625, y: 0.0625 },
  { x: 0.6875, y: 0.0625 },
  { x: 0.6875, y: 0.0625 },
  { x: 0.6875, y: 0.0625 },
  { x: 0.6875, y: 0.0625 },
  { x: 0.6875, y: 0.0625 },
  { x: 0.6875, y: 0.0625 },
  { x: 0.8125, y: 0.0625 },
  { x: 0.8125, y: 0.0625 },
  { x: 0.8125, y: 0.0625 },
  { x: 0.8125, y: 0.0625 },
  { x: 0.8125, y: 0.0625 },
  { x: 0.8125, y: 0.0625 },
  { x: 0.9375, y: 0.0625 },
  { x: 0.9375, y: 0.0625 },
  { x: 0.9375, y: 0.0625 },
  { x: 0.9375, y: 0.0625 },
  { x: 0.9375, y: 0.0625 },
  { x: 0.9375, y: 0.0625 },
  { x: 0.0625, y: 0.1875 },
  { x: 0.0625, y: 0.1875 },
  { x: 0.0625, y: 0.1875 },
  { x: 0.0625, y: 0.1875 },
  { x: 0.0625, y: 0.1875 },
  { x: 0.0625, y: 0.1875 },
  { x: 0.1875, y: 0.1875 },
  { x: 0.1875, y: 0.1875 },
  { x: 0.1875, y: 0.1875 },
  { x: 0.1875, y: 0.1875 },
  { x: 0.1875, y: 0.1875 },
  { x: 0.1875, y: 0.1875 },
  { x: 0.3125, y: 0.1875 },
  { x: 0.3125, y: 0.1875 },
  { x: 0.3125, y: 0.1875 },
  { x: 0.3125, y: 0.1875 },
  { x: 0.3125, y: 0.1875 },
  { x: 0.3125, y: 0.1875 },
  { x: 0.4375, y: 0.1875 },
  { x: 0.4375, y: 0.1875 },
  { x: 0.4375, y: 0.1875 },
  { x: 0.4375, y: 0.1875 },
  { x: 0.4375, y: 0.1875 },
  { x: 0.4375, y: 0.1875 },
  { x: 0.5625, y: 0.1875 },
  { x: 0.5625, y: 0.1875 },
  { x: 0.5625, y: 0.1875 },
  { x: 0.5625, y: 0.1875 },
  { x: 0.5625, y: 0.1875 },
  { x: 0.5625, y: 0.1875 },
  { x: 0.6875, y: 0.1875 },
  { x: 0.6875, y: 0.1875 },
  { x: 0.6875, y: 0.1875 },
  { x: 0.6875, y: 0.1875 },
  { x: 0.6875, y: 0.1875 },
  { x: 0.6875, y: 0.1875 },
  { x: 0.8125, y: 0.1875 },
  { x: 0.8125, y: 0.1875 },
  { x: 0.8125, y: 0.1875 },
  { x: 0.8125, y: 0.1875 },
  { x: 0.8125, y: 0.1875 },
  { x: 0.8125, y: 0.1875 },
  { x: 0.9375, y: 0.1875 },
  { x: 0.9375, y: 0.1875 },
  { x: 0.9375, y: 0.1875 },
  { x: 0.9375, y: 0.1875 },
  { x: 0.9375, y: 0.1875 },
  { x: 0.9375, y: 0.1875 },
  { x: 0.0625, y: 0.3125 },
  { x: 0.0625, y: 0.3125 },
  { x: 0.0625, y: 0.3125 },
  { x: 0.0625, y: 0.3125 },
  { x: 0.0625, y: 0.3125 },
  { x: 0.0625, y: 0.3125 },
  { x: 0.1875, y: 0.3125 },
  { x: 0.1875, y: 0.3125 },
  { x: 0.1875, y: 0.3125 },
  { x: 0.1875, y: 0.3125 },
  { x: 0.1875, y: 0.3125 },
  { x: 0.1875, y: 0.3125 },
  { x: 0.3125, y: 0.3125 },
  { x: 0.3125, y: 0.3125 },
  { x: 0.3125, y: 0.3125 },
  { x: 0.3125, y: 0.3125 },
  { x: 0.3125, y: 0.3125 },
  { x: 0.3125, y: 0.3125 },
  { x: 0.4375, y: 0.3125 },
  { x: 0.4375, y: 0.3125 },
  { x: 0.4375, y: 0.3125 },
  { x: 0.4375, y: 0.3125 },
  { x: 0.4375, y: 0.3125 },
  { x: 0.4375, y: 0.3125 },
  { x: 0.5625, y: 0.3125 },
  { x: 0.5625, y: 0.3125 },
  { x: 0.5625, y: 0.3125 },
  { x: 0.5625, y: 0.3125 },
  { x: 0.5625, y: 0.3125 },
  { x: 0.5625, y: 0.3125 },
  { x: 0.6875, y: 0.3125 },
  { x: 0.6875, y: 0.3125 },
  { x: 0.6875, y: 0.3125 },
  { x: 0.6875, y: 0.3125 },
  { x: 0.6875, y: 0.3125 },
  { x: 0.6875, y: 0.3125 },
  { x: 0.8125, y: 0.3125 },
  { x: 0.8125, y: 0.3125 },
  { x: 0.8125, y: 0.3125 },
  { x: 0.8125, y: 0.3125 },
  { x: 0.8125, y: 0.3125 },
  { x: 0.8125, y: 0.3125 },
  { x: 0.9375, y: 0.3125 },
  { x: 0.9375, y: 0.3125 },
  { x: 0.9375, y: 0.3125 },
  { x: 0.9375, y: 0.3125 },
  { x: 0.9375, y: 0.3125 },
  { x: 0.9375, y: 0.3125 },
  { x: 0.0625, y: 0.4375 },
  { x: 0.0625, y: 0.4375 },
  { x: 0.0625, y: 0.4375 },
  { x: 0.0625, y: 0.4375 },
  { x: 0.0625, y: 0.4375 },
  { x: 0.0625, y: 0.4375 },
  { x: 0.1875, y: 0.4375 },
  { x: 0.1875, y: 0.4375 },
  { x: 0.1875, y: 0.4375 },
  { x: 0.1875, y: 0.4375 },
  { x: 0.1875, y: 0.4375 },
  { x: 0.1875, y: 0.4375 },
  { x: 0.3125, y: 0.4375 },
  { x: 0.3125, y: 0.4375 },
  { x: 0.3125, y: 0.4375 },
  { x: 0.3125, y: 0.4375 },
  { x: 0.3125, y: 0.4375 },
  { x: 0.3125, y: 0.4375 },
  { x: 0.4375, y: 0.4375 },
  { x: 0.4375, y: 0.4375 },
  { x: 0.4375, y: 0.4375 },
  { x: 0.4375, y: 0.4375 },
  { x: 0.4375, y: 0.4375 },
  { x: 0.4375, y: 0.4375 },
  { x: 0.5625, y: 0.4375 },
  { x: 0.5625, y: 0.4375 },
  { x: 0.5625, y: 0.4375 },
  { x: 0.5625, y: 0.4375 },
  { x: 0.5625, y: 0.4375 },
  { x: 0.5625, y: 0.4375 },
  { x: 0.6875, y: 0.4375 },
  { x: 0.6875, y: 0.4375 },
  { x: 0.6875, y: 0.4375 },
  { x: 0.6875, y: 0.4375 },
  { x: 0.6875, y: 0.4375 },
  { x: 0.6875, y: 0.4375 },
  { x: 0.8125, y: 0.4375 },
  { x: 0.8125, y: 0.4375 },
  { x: 0.8125, y: 0.4375 },
  { x: 0.8125, y: 0.4375 },
  { x: 0.8125, y: 0.4375 },
  { x: 0.8125, y: 0.4375 },
  { x: 0.9375, y: 0.4375 },
  { x: 0.9375, y: 0.4375 },
  { x: 0.9375, y: 0.4375 },
  { x: 0.9375, y: 0.4375 },
  { x: 0.9375, y: 0.4375 },
  { x: 0.9375, y: 0.4375 },
  { x: 0.0625, y: 0.5625 },
  { x: 0.0625, y: 0.5625 },
  { x: 0.0625, y: 0.5625 },
  { x: 0.0625, y: 0.5625 },
  { x: 0.0625, y: 0.5625 },
  { x: 0.0625, y: 0.5625 },
  { x: 0.1875, y: 0.5625 },
  { x: 0.1875, y: 0.5625 },
  { x: 0.1875, y: 0.5625 },
  { x: 0.1875, y: 0.5625 },
  { x: 0.1875, y: 0.5625 },
  { x: 0.1875, y: 0.5625 },
  { x: 0.3125, y: 0.5625 },
  { x: 0.3125, y: 0.5625 },
  { x: 0.3125, y: 0.5625 },
  { x: 0.3125, y: 0.5625 },
  { x: 0.3125, y: 0.5625 },
  { x: 0.3125, y: 0.5625 },
  { x: 0.4375, y: 0.5625 },
  { x: 0.4375, y: 0.5625 },
  { x: 0.4375, y: 0.5625 },
  { x: 0.4375, y: 0.5625 },
  { x: 0.4375, y: 0.5625 },
  { x: 0.4375, y: 0.5625 },
  { x: 0.5625, y: 0.5625 },
  { x: 0.5625, y: 0.5625 },
  { x: 0.5625, y: 0.5625 },
  { x: 0.5625, y: 0.5625 },
  { x: 0.5625, y: 0.5625 },
  { x: 0.5625, y: 0.5625 },
  { x: 0.6875, y: 0.5625 },
  { x: 0.6875, y: 0.5625 },
  { x: 0.6875, y: 0.5625 },
  { x: 0.6875, y: 0.5625 },
  { x: 0.6875, y: 0.5625 },
  { x: 0.6875, y: 0.5625 },
  { x: 0.8125, y: 0.5625 },
  { x: 0.8125, y: 0.5625 },
  { x: 0.8125, y: 0.5625 },
  { x: 0.8125, y: 0.5625 },
  { x: 0.8125, y: 0.5625 },
  { x: 0.8125, y: 0.5625 },
  { x: 0.9375, y: 0.5625 },
  { x: 0.9375, y: 0.5625 },
  { x: 0.9375, y: 0.5625 },
  { x: 0.9375, y: 0.5625 },
  { x: 0.9375, y: 0.5625 },
  { x: 0.9375, y: 0.5625 },
  { x: 0.0625, y: 0.6875 },
  { x: 0.0625, y: 0.6875 },
  { x: 0.0625, y: 0.6875 },
  { x: 0.0625, y: 0.6875 },
  { x: 0.0625, y: 0.6875 },
  { x: 0.0625, y: 0.6875 },
  { x: 0.1875, y: 0.6875 },
  { x: 0.1875, y: 0.6875 },
  { x: 0.1875, y: 0.6875 },
  { x: 0.1875, y: 0.6875 },
  { x: 0.1875, y: 0.6875 },
  { x: 0.1875, y: 0.6875 },
  { x: 0.3125, y: 0.6875 },
  { x: 0.3125, y: 0.6875 },
  { x: 0.3125, y: 0.6875 },
  { x: 0.3125, y: 0.6875 },
  { x: 0.3125, y: 0.6875 },
  { x: 0.3125, y: 0.6875 },
  { x: 0.4375, y: 0.6875 },
  { x: 0.4375, y: 0.6875 },
  { x: 0.4375, y: 0.6875 },
  { x: 0.4375, y: 0.6875 },
  { x: 0.4375, y: 0.6875 },
  { x: 0.4375, y: 0.6875 },
  { x: 0.5625, y: 0.6875 },
  { x: 0.5625, y: 0.6875 },
  { x: 0.5625, y: 0.6875 },
  { x: 0.5625, y: 0.6875 },
  { x: 0.5625, y: 0.6875 },
  { x: 0.5625, y: 0.6875 },
  { x: 0.6875, y: 0.6875 },
  { x: 0.6875, y: 0.6875 },
  { x: 0.6875, y: 0.6875 },
  { x: 0.6875, y: 0.6875 },
  { x: 0.6875, y: 0.6875 },
  { x: 0.6875, y: 0.6875 },
  { x: 0.8125, y: 0.6875 },
  { x: 0.8125, y: 0.6875 },
  { x: 0.8125, y: 0.6875 },
  { x: 0.8125, y: 0.6875 },
  { x: 0.8125, y: 0.6875 },
  { x: 0.8125, y: 0.6875 },
  { x: 0.9375, y: 0.6875 },
  { x: 0.9375, y: 0.6875 },
  { x: 0.9375, y: 0.6875 },
  { x: 0.9375, y: 0.6875 },
  { x: 0.9375, y: 0.6875 },
  { x: 0.9375, y: 0.6875 },
  { x: 0.0625, y: 0.8125 },
  { x: 0.0625, y: 0.8125 },
  { x: 0.0625, y: 0.8125 },
  { x: 0.0625, y: 0.8125 },
  { x: 0.0625, y: 0.8125 },
  { x: 0.0625, y: 0.8125 },
  { x: 0.1875, y: 0.8125 },
  { x: 0.1875, y: 0.8125 },
  { x: 0.1875, y: 0.8125 },
  { x: 0.1875, y: 0.8125 },
  { x: 0.1875, y: 0.8125 },
  { x: 0.1875, y: 0.8125 },
  { x: 0.3125, y: 0.8125 },
  { x: 0.3125, y: 0.8125 },
  { x: 0.3125, y: 0.8125 },
  { x: 0.3125, y: 0.8125 },
  { x: 0.3125, y: 0.8125 },
  { x: 0.3125, y: 0.8125 },
  { x: 0.4375, y: 0.8125 },
  { x: 0.4375, y: 0.8125 },
  { x: 0.4375, y: 0.8125 },
  { x: 0.4375, y: 0.8125 },
  { x: 0.4375, y: 0.8125 },
  { x: 0.4375, y: 0.8125 },
  { x: 0.5625, y: 0.8125 },
  { x: 0.5625, y: 0.8125 },
  { x: 0.5625, y: 0.8125 },
  { x: 0.5625, y: 0.8125 },
  { x: 0.5625, y: 0.8125 },
  { x: 0.5625, y: 0.8125 },
  { x: 0.6875, y: 0.8125 },
  { x: 0.6875, y: 0.8125 },
  { x: 0.6875, y: 0.8125 },
  { x: 0.6875, y: 0.8125 },
  { x: 0.6875, y: 0.8125 },
  { x: 0.6875, y: 0.8125 },
  { x: 0.8125, y: 0.8125 },
  { x: 0.8125, y: 0.8125 },
  { x: 0.8125, y: 0.8125 },
  { x: 0.8125, y: 0.8125 },
  { x: 0.8125, y: 0.8125 },
  { x: 0.8125, y: 0.8125 },
  { x: 0.9375, y: 0.8125 },
  { x: 0.9375, y: 0.8125 },
  { x: 0.9375, y: 0.8125 },
  { x: 0.9375, y: 0.8125 },
  { x: 0.9375, y: 0.8125 },
  { x: 0.9375, y: 0.8125 },
  { x: 0.0625, y: 0.9375 },
  { x: 0.0625, y: 0.9375 },
  { x: 0.0625, y: 0.9375 },
  { x: 0.0625, y: 0.9375 },
  { x: 0.0625, y: 0.9375 },
  { x: 0.0625, y: 0.9375 },
  { x: 0.1875, y: 0.9375 },
  { x: 0.1875, y: 0.9375 },
  { x: 0.1875, y: 0.9375 },
  { x: 0.1875, y: 0.9375 },
  { x: 0.1875, y: 0.9375 },
  { x: 0.1875, y: 0.9375 },
  { x: 0.3125, y: 0.9375 },
  { x: 0.3125, y: 0.9375 },
  { x: 0.3125, y: 0.9375 },
  { x: 0.3125, y: 0.9375 },
  { x: 0.3125, y: 0.9375 },
  { x: 0.3125, y: 0.9375 },
  { x: 0.4375, y: 0.9375 },
  { x: 0.4375, y: 0.9375 },
  { x: 0.4375, y: 0.9375 },
  { x: 0.4375, y: 0.9375 },
  { x: 0.4375, y: 0.9375 },
  { x: 0.4375, y: 0.9375 },
  { x: 0.5625, y: 0.9375 },
  { x: 0.5625, y: 0.9375 },
  { x: 0.5625, y: 0.9375 },
  { x: 0.5625, y: 0.9375 },
  { x: 0.5625, y: 0.9375 },
  { x: 0.5625, y: 0.9375 },
  { x: 0.6875, y: 0.9375 },
  { x: 0.6875, y: 0.9375 },
  { x: 0.6875, y: 0.9375 },
  { x: 0.6875, y: 0.9375 },
  { x: 0.6875, y: 0.9375 },
  { x: 0.6875, y: 0.9375 },
  { x: 0.8125, y: 0.9375 },
  { x: 0.8125, y: 0.9375 },
  { x: 0.8125, y: 0.9375 },
  { x: 0.8125, y: 0.9375 },
  { x: 0.8125, y: 0.9375 },
  { x: 0.8125, y: 0.9375 },
  { x: 0.9375, y: 0.9375 },
  { x: 0.9375, y: 0.9375 },
  { x: 0.9375, y: 0.9375 },
  { x: 0.9375, y: 0.9375 },
  { x: 0.9375, y: 0.9375 },
  { x: 0.9375, y: 0.9375 }
];

// src/hand/handposedetector.ts
var HandDetector = class {
  constructor(model18) {
    __publicField(this, "model");
    __publicField(this, "anchors");
    __publicField(this, "anchorsTensor");
    __publicField(this, "inputSize");
    __publicField(this, "inputSizeTensor");
    __publicField(this, "doubleInputSizeTensor");
    this.model = model18;
    this.anchors = anchors2.map((anchor) => [anchor.x, anchor.y]);
    this.anchorsTensor = Zi(this.anchors);
    this.inputSize = this.model && this.model.inputs && this.model.inputs[0].shape ? this.model.inputs[0].shape[2] : 0;
    this.inputSizeTensor = Zt([this.inputSize, this.inputSize]);
    this.doubleInputSizeTensor = Zt([this.inputSize * 2, this.inputSize * 2]);
  }
  normalizeBoxes(boxes) {
    const t = {};
    t.boxOffsets = qe(boxes, [0, 0], [-1, 2]);
    t.boxSizes = qe(boxes, [0, 2], [-1, 2]);
    t.div = xe(t.boxOffsets, this.inputSizeTensor);
    t.boxCenterPoints = ie(t.div, this.anchorsTensor);
    t.halfBoxSizes = xe(t.boxSizes, this.doubleInputSizeTensor);
    t.sub = ge(t.boxCenterPoints, t.halfBoxSizes);
    t.startPoints = V(t.sub, this.inputSizeTensor);
    t.add = ie(t.boxCenterPoints, t.halfBoxSizes);
    t.endPoints = V(t.add, this.inputSizeTensor);
    const res = QE([t.startPoints, t.endPoints], 1);
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    return res;
  }
  normalizeLandmarks(rawPalmLandmarks, index2) {
    const t = {};
    t.reshape = U(rawPalmLandmarks, [-1, 7, 2]);
    t.div = xe(t.reshape, this.inputSizeTensor);
    t.landmarks = ie(t.div, this.anchors[index2]);
    const res = V(t.landmarks, this.inputSizeTensor);
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    return res;
  }
  async predict(input, config3) {
    const t = {};
    t.resize = jn.resizeBilinear(input, [this.inputSize, this.inputSize]);
    t.div = xe(t.resize, constants.tf127);
    t.image = ge(t.div, constants.tf1);
    t.batched = this.model.execute(t.image);
    t.predictions = mr(t.batched);
    t.slice = qe(t.predictions, [0, 0], [-1, 1]);
    t.sigmoid = Hs(t.slice);
    t.scores = mr(t.sigmoid);
    const scores = await t.scores.data();
    t.boxes = qe(t.predictions, [0, 1], [-1, 4]);
    t.norm = this.normalizeBoxes(t.boxes);
    t.nms = await jn.nonMaxSuppressionAsync(t.norm, t.scores, 3 * config3.hand.maxDetected, config3.hand.iouThreshold, config3.hand.minConfidence);
    const nms = await t.nms.array();
    const hands = [];
    for (const index2 of nms) {
      const p = {};
      p.box = qe(t.norm, [index2, 0], [1, -1]);
      p.slice = qe(t.predictions, [index2, 5], [1, 14]);
      p.norm = this.normalizeLandmarks(p.slice, index2);
      p.palmLandmarks = U(p.norm, [-1, 2]);
      const box = await p.box.data();
      const startPoint = box.slice(0, 2);
      const endPoint = box.slice(2, 4);
      const palmLandmarks = await p.palmLandmarks.array();
      const hand3 = { startPoint, endPoint, palmLandmarks, confidence: scores[index2] };
      const scaled = scaleBoxCoordinates2(hand3, [input.shape[2] / this.inputSize, input.shape[1] / this.inputSize]);
      hands.push(scaled);
      Object.keys(p).forEach((tensor) => De(p[tensor]));
    }
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    return hands;
  }
};

// src/hand/handposepipeline.ts
var palmBoxEnlargeFactor = 5;
var handBoxEnlargeFactor = 1.65;
var palmLandmarkIds = [0, 5, 9, 13, 17, 1, 2];
var palmLandmarksPalmBase = 0;
var palmLandmarksMiddleFingerBase = 2;
var lastTime11 = 0;
var HandPipeline = class {
  constructor(handDetector, handPoseModel2) {
    __publicField(this, "handDetector");
    __publicField(this, "handPoseModel");
    __publicField(this, "inputSize");
    __publicField(this, "storedBoxes");
    __publicField(this, "skipped");
    __publicField(this, "detectedHands");
    this.handDetector = handDetector;
    this.handPoseModel = handPoseModel2;
    this.inputSize = this.handPoseModel && this.handPoseModel.inputs[0].shape ? this.handPoseModel.inputs[0].shape[2] : 0;
    this.storedBoxes = [];
    this.skipped = Number.MAX_SAFE_INTEGER;
    this.detectedHands = 0;
  }
  calculateLandmarksBoundingBox(landmarks) {
    const xs2 = landmarks.map((d) => d[0]);
    const ys2 = landmarks.map((d) => d[1]);
    const startPoint = [Math.min(...xs2), Math.min(...ys2)];
    const endPoint = [Math.max(...xs2), Math.max(...ys2)];
    return { startPoint, endPoint };
  }
  getBoxForPalmLandmarks(palmLandmarks, rotationMatrix) {
    const rotatedPalmLandmarks = palmLandmarks.map((coord) => rotatePoint2([...coord, 1], rotationMatrix));
    const boxAroundPalm = this.calculateLandmarksBoundingBox(rotatedPalmLandmarks);
    return enlargeBox2(squarifyBox2(boxAroundPalm), palmBoxEnlargeFactor);
  }
  getBoxForHandLandmarks(landmarks) {
    const boundingBox = this.calculateLandmarksBoundingBox(landmarks);
    const boxAroundHand = enlargeBox2(squarifyBox2(boundingBox), handBoxEnlargeFactor);
    boxAroundHand.palmLandmarks = [];
    for (let i = 0; i < palmLandmarkIds.length; i++) {
      boxAroundHand.palmLandmarks.push(landmarks[palmLandmarkIds[i]].slice(0, 2));
    }
    return boxAroundHand;
  }
  transformRawCoords(rawCoords, box2, angle, rotationMatrix) {
    const boxSize = getBoxSize2(box2);
    const scaleFactor = [boxSize[0] / this.inputSize, boxSize[1] / this.inputSize, (boxSize[0] + boxSize[1]) / this.inputSize / 2];
    const coordsScaled = rawCoords.map((coord) => [
      scaleFactor[0] * (coord[0] - this.inputSize / 2),
      scaleFactor[1] * (coord[1] - this.inputSize / 2),
      scaleFactor[2] * coord[2]
    ]);
    const coordsRotationMatrix = buildRotationMatrix2(angle, [0, 0]);
    const coordsRotated = coordsScaled.map((coord) => {
      const rotated = rotatePoint2(coord, coordsRotationMatrix);
      return [...rotated, coord[2]];
    });
    const inverseRotationMatrix = invertTransformMatrix2(rotationMatrix);
    const boxCenter = [...getBoxCenter2(box2), 1];
    const originalBoxCenter = [
      dot2(boxCenter, inverseRotationMatrix[0]),
      dot2(boxCenter, inverseRotationMatrix[1])
    ];
    return coordsRotated.map((coord) => [
      Math.trunc(coord[0] + originalBoxCenter[0]),
      Math.trunc(coord[1] + originalBoxCenter[1]),
      Math.trunc(coord[2])
    ]);
  }
  async estimateHands(image, config3) {
    let useFreshBox = false;
    let boxes;
    const skipTime = (config3.hand.skipTime || 0) > now() - lastTime11;
    const skipFrame = this.skipped < (config3.hand.skipFrames || 0);
    if (config3.skipAllowed && skipTime && skipFrame) {
      boxes = await this.handDetector.predict(image, config3);
      this.skipped = 0;
    }
    if (config3.skipAllowed)
      this.skipped++;
    if (boxes && boxes.length > 0 && (boxes.length !== this.detectedHands && this.detectedHands !== config3.hand.maxDetected || !config3.hand.landmarks)) {
      this.detectedHands = 0;
      this.storedBoxes = [...boxes];
      if (this.storedBoxes.length > 0)
        useFreshBox = true;
    }
    const hands = [];
    for (let i = 0; i < this.storedBoxes.length; i++) {
      const currentBox = this.storedBoxes[i];
      if (!currentBox)
        continue;
      if (config3.hand.landmarks) {
        const angle = config3.hand.rotation ? computeRotation2(currentBox.palmLandmarks[palmLandmarksPalmBase], currentBox.palmLandmarks[palmLandmarksMiddleFingerBase]) : 0;
        const palmCenter = getBoxCenter2(currentBox);
        const palmCenterNormalized = [palmCenter[0] / image.shape[2], palmCenter[1] / image.shape[1]];
        const rotatedImage = config3.hand.rotation && env.kernels.includes("rotatewithoffset") ? jn.rotateWithOffset(image, angle, 0, palmCenterNormalized) : image.clone();
        const rotationMatrix = buildRotationMatrix2(-angle, palmCenter);
        const newBox = useFreshBox ? this.getBoxForPalmLandmarks(currentBox.palmLandmarks, rotationMatrix) : currentBox;
        const croppedInput = cutBoxFromImageAndResize(newBox, rotatedImage, [this.inputSize, this.inputSize]);
        const handImage = xe(croppedInput, constants.tf255);
        De(croppedInput);
        De(rotatedImage);
        const [confidenceT, keypoints] = this.handPoseModel.execute(handImage);
        lastTime11 = now();
        De(handImage);
        const confidence = (await confidenceT.data())[0];
        De(confidenceT);
        if (confidence >= config3.hand.minConfidence / 4) {
          const keypointsReshaped = U(keypoints, [-1, 3]);
          const rawCoords = await keypointsReshaped.array();
          De(keypoints);
          De(keypointsReshaped);
          const coords = this.transformRawCoords(rawCoords, newBox, angle, rotationMatrix);
          const nextBoundingBox = this.getBoxForHandLandmarks(coords);
          this.storedBoxes[i] = { ...nextBoundingBox, confidence };
          const result = {
            landmarks: coords,
            confidence,
            boxConfidence: currentBox.confidence,
            fingerConfidence: confidence,
            box: { topLeft: nextBoundingBox.startPoint, bottomRight: nextBoundingBox.endPoint }
          };
          hands.push(result);
        } else {
          this.storedBoxes[i] = null;
        }
        De(keypoints);
      } else {
        const enlarged = enlargeBox2(squarifyBox2(currentBox), handBoxEnlargeFactor);
        const result = {
          confidence: currentBox.confidence,
          boxConfidence: currentBox.confidence,
          fingerConfidence: 0,
          box: { topLeft: enlarged.startPoint, bottomRight: enlarged.endPoint },
          landmarks: []
        };
        hands.push(result);
      }
    }
    this.storedBoxes = this.storedBoxes.filter((a) => a !== null);
    this.detectedHands = hands.length;
    if (hands.length > config3.hand.maxDetected)
      hands.length = config3.hand.maxDetected;
    return hands;
  }
};

// src/hand/fingerdef.ts
var Finger = {
  thumb: 0,
  index: 1,
  middle: 2,
  ring: 3,
  pinky: 4,
  all: [0, 1, 2, 3, 4],
  nameMapping: { 0: "thumb", 1: "index", 2: "middle", 3: "ring", 4: "pinky" },
  pointsMapping: {
    0: [[0, 1], [1, 2], [2, 3], [3, 4]],
    1: [[0, 5], [5, 6], [6, 7], [7, 8]],
    2: [[0, 9], [9, 10], [10, 11], [11, 12]],
    3: [[0, 13], [13, 14], [14, 15], [15, 16]],
    4: [[0, 17], [17, 18], [18, 19], [19, 20]]
  },
  getName: (value) => Finger.nameMapping[value],
  getPoints: (value) => Finger.pointsMapping[value]
};
var FingerCurl = {
  none: 0,
  half: 1,
  full: 2,
  nameMapping: { 0: "none", 1: "half", 2: "full" },
  getName: (value) => FingerCurl.nameMapping[value]
};
var FingerDirection = {
  verticalUp: 0,
  verticalDown: 1,
  horizontalLeft: 2,
  horizontalRight: 3,
  diagonalUpRight: 4,
  diagonalUpLeft: 5,
  diagonalDownRight: 6,
  diagonalDownLeft: 7,
  nameMapping: { 0: "verticalUp", 1: "verticalDown", 2: "horizontalLeft", 3: "horizontalRight", 4: "diagonalUpRight", 5: "diagonalUpLeft", 6: "diagonalDownRight", 7: "diagonalDownLeft" },
  getName: (value) => FingerDirection.nameMapping[value]
};
var FingerGesture = class {
  constructor(name) {
    __publicField(this, "name");
    __publicField(this, "curls");
    __publicField(this, "directions");
    __publicField(this, "weights");
    __publicField(this, "weightsRelative");
    this.name = name;
    this.curls = {};
    this.directions = {};
    this.weights = [1, 1, 1, 1, 1];
    this.weightsRelative = [1, 1, 1, 1, 1];
  }
  curl(finger, curl, confidence) {
    if (typeof this.curls[finger] === "undefined")
      this.curls[finger] = [];
    this.curls[finger].push([curl, confidence]);
  }
  direction(finger, position, confidence) {
    if (!this.directions[finger])
      this.directions[finger] = [];
    this.directions[finger].push([position, confidence]);
  }
  weight(finger, weight) {
    this.weights[finger] = weight;
    const total = this.weights.reduce((a, b) => a + b, 0);
    this.weightsRelative = this.weights.map((el2) => el2 * 5 / total);
  }
  matchAgainst(detectedCurls, detectedDirections) {
    let confidence = 0;
    for (const fingerIdx in detectedCurls) {
      const detectedCurl = detectedCurls[fingerIdx];
      const expectedCurls = this.curls[fingerIdx];
      if (typeof expectedCurls === "undefined") {
        confidence += this.weightsRelative[fingerIdx];
        continue;
      }
      for (const [expectedCurl, score] of expectedCurls) {
        if (detectedCurl === expectedCurl) {
          confidence += score * this.weightsRelative[fingerIdx];
          break;
        }
      }
    }
    for (const fingerIdx in detectedDirections) {
      const detectedDirection = detectedDirections[fingerIdx];
      const expectedDirections = this.directions[fingerIdx];
      if (typeof expectedDirections === "undefined") {
        confidence += this.weightsRelative[fingerIdx];
        continue;
      }
      for (const [expectedDirection, score] of expectedDirections) {
        if (detectedDirection === expectedDirection) {
          confidence += score * this.weightsRelative[fingerIdx];
          break;
        }
      }
    }
    return confidence / 10;
  }
};

// src/hand/fingergesture.ts
var { thumb, index, middle, ring, pinky } = Finger;
var { none, half, full } = FingerCurl;
var { verticalUp, verticalDown, horizontalLeft, horizontalRight, diagonalUpRight, diagonalUpLeft, diagonalDownRight, diagonalDownLeft } = FingerDirection;
var ThumbsUp = new FingerGesture("thumbs up");
ThumbsUp.curl(thumb, none, 1);
ThumbsUp.direction(thumb, verticalUp, 1);
ThumbsUp.direction(thumb, diagonalUpLeft, 0.25);
ThumbsUp.direction(thumb, diagonalUpRight, 0.25);
for (const finger of [Finger.index, Finger.middle, Finger.ring, Finger.pinky]) {
  ThumbsUp.curl(finger, full, 1);
  ThumbsUp.direction(finger, horizontalLeft, 1);
  ThumbsUp.direction(finger, horizontalRight, 1);
}
var Victory = new FingerGesture("victory");
Victory.curl(thumb, half, 0.5);
Victory.curl(thumb, none, 0.5);
Victory.direction(thumb, verticalUp, 1);
Victory.direction(thumb, diagonalUpLeft, 1);
Victory.curl(index, none, 1);
Victory.direction(index, verticalUp, 0.75);
Victory.direction(index, diagonalUpLeft, 1);
Victory.curl(middle, none, 1);
Victory.direction(middle, verticalUp, 1);
Victory.direction(middle, diagonalUpLeft, 0.75);
Victory.curl(ring, full, 1);
Victory.direction(ring, verticalUp, 0.2);
Victory.direction(ring, diagonalUpLeft, 1);
Victory.direction(ring, horizontalLeft, 0.2);
Victory.curl(pinky, full, 1);
Victory.direction(pinky, verticalUp, 0.2);
Victory.direction(pinky, diagonalUpLeft, 1);
Victory.direction(pinky, horizontalLeft, 0.2);
Victory.weight(index, 2);
Victory.weight(middle, 2);
var Point = new FingerGesture("point");
Point.curl(thumb, full, 1);
Point.curl(index, none, 0.5);
Point.curl(middle, full, 0.5);
Point.curl(ring, full, 0.5);
Point.curl(pinky, full, 0.5);
Point.weight(index, 2);
Point.weight(middle, 2);
var MiddleFinger = new FingerGesture("middle finger");
MiddleFinger.curl(thumb, none, 1);
MiddleFinger.curl(index, full, 0.5);
MiddleFinger.curl(middle, full, 0.5);
MiddleFinger.curl(ring, full, 0.5);
MiddleFinger.curl(pinky, full, 0.5);
MiddleFinger.weight(index, 2);
MiddleFinger.weight(middle, 2);
var OpenPalm = new FingerGesture("open palm");
OpenPalm.curl(thumb, none, 0.75);
OpenPalm.curl(index, none, 0.75);
OpenPalm.curl(middle, none, 0.75);
OpenPalm.curl(ring, none, 0.75);
OpenPalm.curl(pinky, none, 0.75);
var fingergesture_default = [ThumbsUp, Victory, Point, MiddleFinger, OpenPalm];

// src/hand/fingerpose.ts
var minConfidence = 0.7;
var options2 = {
  HALF_CURL_START_LIMIT: 60,
  NO_CURL_START_LIMIT: 130,
  DISTANCE_VOTE_POWER: 1.1,
  SINGLE_ANGLE_VOTE_POWER: 0.9,
  TOTAL_ANGLE_VOTE_POWER: 1.6
};
function calculateSlope(point1x, point1y, point2x, point2y) {
  const value = (point1y - point2y) / (point1x - point2x);
  let slope = Math.atan(value) * 180 / Math.PI;
  if (slope <= 0)
    slope = -slope;
  else if (slope > 0)
    slope = 180 - slope;
  return slope;
}
function getSlopes(point1, point2) {
  if (!point1 || !point2)
    return [0, 0];
  const slopeXY = calculateSlope(point1[0], point1[1], point2[0], point2[1]);
  if (point1.length === 2)
    return slopeXY;
  const slopeYZ = calculateSlope(point1[1], point1[2], point2[1], point2[2]);
  return [slopeXY, slopeYZ];
}
function angleOrientationAt(angle, weightageAt = 1) {
  let isVertical = 0;
  let isDiagonal = 0;
  let isHorizontal = 0;
  if (angle >= 75 && angle <= 105)
    isVertical = 1 * weightageAt;
  else if (angle >= 25 && angle <= 155)
    isDiagonal = 1 * weightageAt;
  else
    isHorizontal = 1 * weightageAt;
  return [isVertical, isDiagonal, isHorizontal];
}
function estimateFingerCurl(startPoint, midPoint, endPoint) {
  const start_mid_x_dist = startPoint[0] - midPoint[0];
  const start_end_x_dist = startPoint[0] - endPoint[0];
  const mid_end_x_dist = midPoint[0] - endPoint[0];
  const start_mid_y_dist = startPoint[1] - midPoint[1];
  const start_end_y_dist = startPoint[1] - endPoint[1];
  const mid_end_y_dist = midPoint[1] - endPoint[1];
  const start_mid_z_dist = startPoint[2] - midPoint[2];
  const start_end_z_dist = startPoint[2] - endPoint[2];
  const mid_end_z_dist = midPoint[2] - endPoint[2];
  const start_mid_dist = Math.sqrt(start_mid_x_dist * start_mid_x_dist + start_mid_y_dist * start_mid_y_dist + start_mid_z_dist * start_mid_z_dist);
  const start_end_dist = Math.sqrt(start_end_x_dist * start_end_x_dist + start_end_y_dist * start_end_y_dist + start_end_z_dist * start_end_z_dist);
  const mid_end_dist = Math.sqrt(mid_end_x_dist * mid_end_x_dist + mid_end_y_dist * mid_end_y_dist + mid_end_z_dist * mid_end_z_dist);
  let cos_in = (mid_end_dist * mid_end_dist + start_mid_dist * start_mid_dist - start_end_dist * start_end_dist) / (2 * mid_end_dist * start_mid_dist);
  if (cos_in > 1)
    cos_in = 1;
  else if (cos_in < -1)
    cos_in = -1;
  let angleOfCurve = Math.acos(cos_in);
  angleOfCurve = 57.2958 * angleOfCurve % 180;
  let fingerCurl;
  if (angleOfCurve > options2.NO_CURL_START_LIMIT)
    fingerCurl = FingerCurl.none;
  else if (angleOfCurve > options2.HALF_CURL_START_LIMIT)
    fingerCurl = FingerCurl.half;
  else
    fingerCurl = FingerCurl.full;
  return fingerCurl;
}
function estimateHorizontalDirection(start_end_x_dist, start_mid_x_dist, mid_end_x_dist, max_dist_x) {
  let estimatedDirection;
  if (max_dist_x === Math.abs(start_end_x_dist)) {
    if (start_end_x_dist > 0)
      estimatedDirection = FingerDirection.horizontalLeft;
    else
      estimatedDirection = FingerDirection.horizontalRight;
  } else if (max_dist_x === Math.abs(start_mid_x_dist)) {
    if (start_mid_x_dist > 0)
      estimatedDirection = FingerDirection.horizontalLeft;
    else
      estimatedDirection = FingerDirection.horizontalRight;
  } else {
    if (mid_end_x_dist > 0)
      estimatedDirection = FingerDirection.horizontalLeft;
    else
      estimatedDirection = FingerDirection.horizontalRight;
  }
  return estimatedDirection;
}
function estimateVerticalDirection(start_end_y_dist, start_mid_y_dist, mid_end_y_dist, max_dist_y) {
  let estimatedDirection;
  if (max_dist_y === Math.abs(start_end_y_dist)) {
    if (start_end_y_dist < 0)
      estimatedDirection = FingerDirection.verticalDown;
    else
      estimatedDirection = FingerDirection.verticalUp;
  } else if (max_dist_y === Math.abs(start_mid_y_dist)) {
    if (start_mid_y_dist < 0)
      estimatedDirection = FingerDirection.verticalDown;
    else
      estimatedDirection = FingerDirection.verticalUp;
  } else {
    if (mid_end_y_dist < 0)
      estimatedDirection = FingerDirection.verticalDown;
    else
      estimatedDirection = FingerDirection.verticalUp;
  }
  return estimatedDirection;
}
function estimateDiagonalDirection(start_end_y_dist, start_mid_y_dist, mid_end_y_dist, max_dist_y, start_end_x_dist, start_mid_x_dist, mid_end_x_dist, max_dist_x) {
  let estimatedDirection;
  const reqd_vertical_direction = estimateVerticalDirection(start_end_y_dist, start_mid_y_dist, mid_end_y_dist, max_dist_y);
  const reqd_horizontal_direction = estimateHorizontalDirection(start_end_x_dist, start_mid_x_dist, mid_end_x_dist, max_dist_x);
  if (reqd_vertical_direction === FingerDirection.verticalUp) {
    if (reqd_horizontal_direction === FingerDirection.horizontalLeft)
      estimatedDirection = FingerDirection.diagonalUpLeft;
    else
      estimatedDirection = FingerDirection.diagonalUpRight;
  } else {
    if (reqd_horizontal_direction === FingerDirection.horizontalLeft)
      estimatedDirection = FingerDirection.diagonalDownLeft;
    else
      estimatedDirection = FingerDirection.diagonalDownRight;
  }
  return estimatedDirection;
}
function calculateFingerDirection(startPoint, midPoint, endPoint, fingerSlopes) {
  const start_mid_x_dist = startPoint[0] - midPoint[0];
  const start_end_x_dist = startPoint[0] - endPoint[0];
  const mid_end_x_dist = midPoint[0] - endPoint[0];
  const start_mid_y_dist = startPoint[1] - midPoint[1];
  const start_end_y_dist = startPoint[1] - endPoint[1];
  const mid_end_y_dist = midPoint[1] - endPoint[1];
  const max_dist_x = Math.max(Math.abs(start_mid_x_dist), Math.abs(start_end_x_dist), Math.abs(mid_end_x_dist));
  const max_dist_y = Math.max(Math.abs(start_mid_y_dist), Math.abs(start_end_y_dist), Math.abs(mid_end_y_dist));
  let voteVertical = 0;
  let voteDiagonal = 0;
  let voteHorizontal = 0;
  const start_end_x_y_dist_ratio = max_dist_y / (max_dist_x + 1e-5);
  if (start_end_x_y_dist_ratio > 1.5)
    voteVertical += options2.DISTANCE_VOTE_POWER;
  else if (start_end_x_y_dist_ratio > 0.66)
    voteDiagonal += options2.DISTANCE_VOTE_POWER;
  else
    voteHorizontal += options2.DISTANCE_VOTE_POWER;
  const start_mid_dist = Math.sqrt(start_mid_x_dist * start_mid_x_dist + start_mid_y_dist * start_mid_y_dist);
  const start_end_dist = Math.sqrt(start_end_x_dist * start_end_x_dist + start_end_y_dist * start_end_y_dist);
  const mid_end_dist = Math.sqrt(mid_end_x_dist * mid_end_x_dist + mid_end_y_dist * mid_end_y_dist);
  const max_dist = Math.max(start_mid_dist, start_end_dist, mid_end_dist);
  let calc_start_point_x = startPoint[0];
  let calc_start_point_y = startPoint[1];
  let calc_end_point_x = endPoint[0];
  let calc_end_point_y = endPoint[1];
  if (max_dist === start_mid_dist) {
    calc_end_point_x = endPoint[0];
    calc_end_point_y = endPoint[1];
  } else if (max_dist === mid_end_dist) {
    calc_start_point_x = midPoint[0];
    calc_start_point_y = midPoint[1];
  }
  const calcStartPoint = [calc_start_point_x, calc_start_point_y];
  const calcEndPoint = [calc_end_point_x, calc_end_point_y];
  const totalAngle = getSlopes(calcStartPoint, calcEndPoint);
  const votes = angleOrientationAt(totalAngle, options2.TOTAL_ANGLE_VOTE_POWER);
  voteVertical += votes[0];
  voteDiagonal += votes[1];
  voteHorizontal += votes[2];
  for (const fingerSlope of fingerSlopes) {
    const fingerVotes = angleOrientationAt(fingerSlope, options2.SINGLE_ANGLE_VOTE_POWER);
    voteVertical += fingerVotes[0];
    voteDiagonal += fingerVotes[1];
    voteHorizontal += fingerVotes[2];
  }
  let estimatedDirection;
  if (voteVertical === Math.max(voteVertical, voteDiagonal, voteHorizontal)) {
    estimatedDirection = estimateVerticalDirection(start_end_y_dist, start_mid_y_dist, mid_end_y_dist, max_dist_y);
  } else if (voteHorizontal === Math.max(voteDiagonal, voteHorizontal)) {
    estimatedDirection = estimateHorizontalDirection(start_end_x_dist, start_mid_x_dist, mid_end_x_dist, max_dist_x);
  } else {
    estimatedDirection = estimateDiagonalDirection(start_end_y_dist, start_mid_y_dist, mid_end_y_dist, max_dist_y, start_end_x_dist, start_mid_x_dist, mid_end_x_dist, max_dist_x);
  }
  return estimatedDirection;
}
function estimate(landmarks) {
  const slopesXY = [];
  const slopesYZ = [];
  const fingerCurls = [];
  const fingerDirections = [];
  if (!landmarks)
    return { curls: fingerCurls, directions: fingerDirections };
  for (const finger of Finger.all) {
    const points = Finger.getPoints(finger);
    const slopeAtXY = [];
    const slopeAtYZ = [];
    for (const point2 of points) {
      const point1 = landmarks[point2[0]];
      const point22 = landmarks[point2[1]];
      const slopes = getSlopes(point1, point22);
      const slopeXY = slopes[0];
      const slopeYZ = slopes[1];
      slopeAtXY.push(slopeXY);
      slopeAtYZ.push(slopeYZ);
    }
    slopesXY.push(slopeAtXY);
    slopesYZ.push(slopeAtYZ);
  }
  for (const finger of Finger.all) {
    const pointIndexAt = finger === Finger.thumb ? 1 : 0;
    const fingerPointsAt = Finger.getPoints(finger);
    const startPoint = landmarks[fingerPointsAt[pointIndexAt][0]];
    const midPoint = landmarks[fingerPointsAt[pointIndexAt + 1][1]];
    const endPoint = landmarks[fingerPointsAt[3][1]];
    const fingerCurled = estimateFingerCurl(startPoint, midPoint, endPoint);
    const fingerPosition = calculateFingerDirection(startPoint, midPoint, endPoint, slopesXY[finger].slice(pointIndexAt));
    fingerCurls[finger] = fingerCurled;
    fingerDirections[finger] = fingerPosition;
  }
  return { curls: fingerCurls, directions: fingerDirections };
}
function analyze(keypoints) {
  if (!keypoints || keypoints.length === 0)
    return null;
  const estimatorRes = estimate(keypoints);
  const landmarks = {};
  for (const fingerIdx of Finger.all) {
    landmarks[Finger.getName(fingerIdx)] = {
      curl: FingerCurl.getName(estimatorRes.curls[fingerIdx]),
      direction: FingerDirection.getName(estimatorRes.directions[fingerIdx])
    };
  }
  return landmarks;
}
function match(keypoints) {
  const poses = [];
  if (!keypoints || keypoints.length === 0)
    return poses;
  const estimatorRes = estimate(keypoints);
  for (const gesture2 of fingergesture_default) {
    const confidence = gesture2.matchAgainst(estimatorRes.curls, estimatorRes.directions);
    if (confidence >= minConfidence)
      poses.push({ name: gesture2.name, confidence });
  }
  return poses;
}

// src/hand/handpose.ts
var meshAnnotations2 = {
  thumb: [1, 2, 3, 4],
  index: [5, 6, 7, 8],
  middle: [9, 10, 11, 12],
  ring: [13, 14, 15, 16],
  pinky: [17, 18, 19, 20],
  palm: [0]
};
var handDetectorModel;
var handPoseModel;
var handPipeline;
async function predict12(input, config3) {
  const predictions = await handPipeline.estimateHands(input, config3);
  if (!predictions)
    return [];
  const hands = [];
  for (let i = 0; i < predictions.length; i++) {
    const annotations2 = {};
    if (predictions[i].landmarks) {
      for (const key of Object.keys(meshAnnotations2)) {
        annotations2[key] = meshAnnotations2[key].map((index2) => predictions[i].landmarks[index2]);
      }
    }
    const keypoints = predictions[i].landmarks;
    let box = [Number.MAX_SAFE_INTEGER, Number.MAX_SAFE_INTEGER, 0, 0];
    let boxRaw = [0, 0, 0, 0];
    if (keypoints && keypoints.length > 0) {
      for (const pt2 of keypoints) {
        if (pt2[0] < box[0])
          box[0] = pt2[0];
        if (pt2[1] < box[1])
          box[1] = pt2[1];
        if (pt2[0] > box[2])
          box[2] = pt2[0];
        if (pt2[1] > box[3])
          box[3] = pt2[1];
      }
      box[2] -= box[0];
      box[3] -= box[1];
      boxRaw = [box[0] / (input.shape[2] || 0), box[1] / (input.shape[1] || 0), box[2] / (input.shape[2] || 0), box[3] / (input.shape[1] || 0)];
    } else {
      box = predictions[i].box ? [
        Math.trunc(Math.max(0, predictions[i].box.topLeft[0])),
        Math.trunc(Math.max(0, predictions[i].box.topLeft[1])),
        Math.trunc(Math.min(input.shape[2] || 0, predictions[i].box.bottomRight[0]) - Math.max(0, predictions[i].box.topLeft[0])),
        Math.trunc(Math.min(input.shape[1] || 0, predictions[i].box.bottomRight[1]) - Math.max(0, predictions[i].box.topLeft[1]))
      ] : [0, 0, 0, 0];
      boxRaw = [
        predictions[i].box.topLeft[0] / (input.shape[2] || 0),
        predictions[i].box.topLeft[1] / (input.shape[1] || 0),
        (predictions[i].box.bottomRight[0] - predictions[i].box.topLeft[0]) / (input.shape[2] || 0),
        (predictions[i].box.bottomRight[1] - predictions[i].box.topLeft[1]) / (input.shape[1] || 0)
      ];
    }
    const landmarks = analyze(keypoints);
    hands.push({
      id: i,
      score: Math.round(100 * predictions[i].confidence) / 100,
      boxScore: Math.round(100 * predictions[i].boxConfidence) / 100,
      fingerScore: Math.round(100 * predictions[i].fingerConfidence) / 100,
      label: "hand",
      box,
      boxRaw,
      keypoints,
      annotations: annotations2,
      landmarks
    });
  }
  return hands;
}
async function load13(config3) {
  var _a2, _b2;
  if (env.initial) {
    handDetectorModel = null;
    handPoseModel = null;
  }
  if (!handDetectorModel || !handPoseModel) {
    [handDetectorModel, handPoseModel] = await Promise.all([
      config3.hand.enabled ? loadModel((_a2 = config3.hand.detector) == null ? void 0 : _a2.modelPath) : null,
      config3.hand.landmarks ? loadModel((_b2 = config3.hand.skeleton) == null ? void 0 : _b2.modelPath) : null
    ]);
  } else {
    if (config3.debug)
      log("cached model:", handDetectorModel["modelUrl"]);
    if (config3.debug)
      log("cached model:", handPoseModel["modelUrl"]);
  }
  const handDetector = new HandDetector(handDetectorModel);
  handPipeline = new HandPipeline(handDetector, handPoseModel);
  return [handDetectorModel, handPoseModel];
}

// src/hand/handtrack.ts
var models2 = [null, null];
var modelOutputNodes = ["StatefulPartitionedCall/Postprocessor/Slice", "StatefulPartitionedCall/Postprocessor/ExpandDims_1"];
var inputSize7 = [[0, 0], [0, 0]];
var classes = ["hand", "fist", "pinch", "point", "face", "tip", "pinchtip"];
var faceIndex = 4;
var boxExpandFact = 1.6;
var maxDetectorResolution = 512;
var detectorExpandFact = 1.4;
var skipped11 = Number.MAX_SAFE_INTEGER;
var lastTime12 = 0;
var outputSize = [0, 0];
var cache4 = {
  boxes: [],
  hands: []
};
var fingerMap = {
  thumb: [1, 2, 3, 4],
  index: [5, 6, 7, 8],
  middle: [9, 10, 11, 12],
  ring: [13, 14, 15, 16],
  pinky: [17, 18, 19, 20],
  base: [0],
  palm: [0, 17, 13, 9, 5, 1, 0]
};
async function loadDetect2(config3) {
  var _a2;
  if (env.initial)
    models2[0] = null;
  if (!models2[0]) {
    fakeOps(["tensorlistreserve", "enter", "tensorlistfromtensor", "merge", "loopcond", "switch", "exit", "tensorliststack", "nextiteration", "tensorlistsetitem", "tensorlistgetitem", "reciprocal", "shape", "split", "where"], config3);
    models2[0] = await loadModel((_a2 = config3.hand.detector) == null ? void 0 : _a2.modelPath);
    const inputs = Object.values(models2[0].modelSignature["inputs"]);
    inputSize7[0][0] = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[1].size) : 0;
    inputSize7[0][1] = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[2].size) : 0;
  } else if (config3.debug)
    log("cached model:", models2[0]["modelUrl"]);
  return models2[0];
}
async function loadSkeleton(config3) {
  var _a2;
  if (env.initial)
    models2[1] = null;
  if (!models2[1]) {
    models2[1] = await loadModel((_a2 = config3.hand.skeleton) == null ? void 0 : _a2.modelPath);
    const inputs = Object.values(models2[1].modelSignature["inputs"]);
    inputSize7[1][0] = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[1].size) : 0;
    inputSize7[1][1] = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[2].size) : 0;
  } else if (config3.debug)
    log("cached model:", models2[1]["modelUrl"]);
  return models2[1];
}
async function detectHands(input, config3) {
  const hands = [];
  if (!input || !models2[0])
    return hands;
  const t = {};
  const ratio = (input.shape[2] || 1) / (input.shape[1] || 1);
  const height = Math.min(Math.round((input.shape[1] || 0) / 8) * 8, maxDetectorResolution);
  const width = Math.round(height * ratio / 8) * 8;
  t.resize = jn.resizeBilinear(input, [height, width]);
  t.cast = le(t.resize, "int32");
  [t.rawScores, t.rawBoxes] = await models2[0].executeAsync(t.cast, modelOutputNodes);
  t.boxes = mr(t.rawBoxes, [0, 2]);
  t.scores = mr(t.rawScores, [0]);
  const classScores = Fs(t.scores, 1);
  De(classScores[faceIndex]);
  classScores.splice(faceIndex, 1);
  t.filtered = es(classScores, 1);
  De(classScores);
  t.max = As(t.filtered, 1);
  t.argmax = Xu(t.filtered, 1);
  let id2 = 0;
  t.nms = await jn.nonMaxSuppressionAsync(t.boxes, t.max, (config3.hand.maxDetected || 0) + 1, config3.hand.iouThreshold || 0, config3.hand.minConfidence || 1);
  const nms = await t.nms.data();
  const scores = await t.max.data();
  const classNum = await t.argmax.data();
  for (const nmsIndex of Array.from(nms)) {
    const boxSlice = qe(t.boxes, nmsIndex, 1);
    const boxYX = await boxSlice.data();
    De(boxSlice);
    const boxData = [boxYX[1], boxYX[0], boxYX[3] - boxYX[1], boxYX[2] - boxYX[0]];
    const boxRaw = scale(boxData, detectorExpandFact);
    const boxFull = [Math.trunc(boxData[0] * outputSize[0]), Math.trunc(boxData[1] * outputSize[1]), Math.trunc(boxData[2] * outputSize[0]), Math.trunc(boxData[3] * outputSize[1])];
    const score = scores[nmsIndex];
    const label = classes[classNum[nmsIndex]];
    const hand3 = { id: id2++, score, box: boxFull, boxRaw, label };
    hands.push(hand3);
  }
  Object.keys(t).forEach((tensor) => De(t[tensor]));
  hands.sort((a, b) => b.score - a.score);
  if (hands.length > (config3.hand.maxDetected || 1))
    hands.length = config3.hand.maxDetected || 1;
  return hands;
}
async function detectFingers(input, h, config3) {
  const hand3 = {
    id: h.id,
    score: Math.round(100 * h.score) / 100,
    boxScore: Math.round(100 * h.score) / 100,
    fingerScore: 0,
    box: h.box,
    boxRaw: h.boxRaw,
    label: h.label,
    keypoints: [],
    landmarks: {},
    annotations: {}
  };
  if (input && models2[1] && config3.hand.landmarks && h.score > (config3.hand.minConfidence || 0)) {
    const t = {};
    const boxCrop = [h.boxRaw[1], h.boxRaw[0], h.boxRaw[3] + h.boxRaw[1], h.boxRaw[2] + h.boxRaw[0]];
    t.crop = jn.cropAndResize(input, [boxCrop], [0], [inputSize7[1][0], inputSize7[1][1]], "bilinear");
    t.div = xe(t.crop, constants.tf255);
    [t.score, t.keypoints] = models2[1].execute(t.div, ["Identity_1", "Identity"]);
    const rawScore = (await t.score.data())[0];
    const score = (100 - Math.trunc(100 / (1 + Math.exp(rawScore)))) / 100;
    if (score >= (config3.hand.minConfidence || 0)) {
      hand3.fingerScore = score;
      t.reshaped = U(t.keypoints, [-1, 3]);
      const coordsData = await t.reshaped.array();
      const coordsRaw = coordsData.map((kpt4) => [kpt4[0] / inputSize7[1][1], kpt4[1] / inputSize7[1][0], kpt4[2] || 0]);
      const coordsNorm = coordsRaw.map((kpt4) => [kpt4[0] * h.boxRaw[2], kpt4[1] * h.boxRaw[3], kpt4[2] || 0]);
      hand3.keypoints = coordsNorm.map((kpt4) => [outputSize[0] * (kpt4[0] + h.boxRaw[0]), outputSize[1] * (kpt4[1] + h.boxRaw[1]), kpt4[2] || 0]);
      hand3.landmarks = analyze(hand3.keypoints);
      for (const key of Object.keys(fingerMap)) {
        hand3.annotations[key] = fingerMap[key].map((index2) => hand3.landmarks && hand3.keypoints[index2] ? hand3.keypoints[index2] : null);
      }
    }
    Object.keys(t).forEach((tensor) => De(t[tensor]));
  }
  return hand3;
}
async function predict13(input, config3) {
  var _a2, _b2;
  if (!models2[0] || !models2[1] || !((_a2 = models2[0]) == null ? void 0 : _a2.inputs[0].shape) || !((_b2 = models2[1]) == null ? void 0 : _b2.inputs[0].shape))
    return [];
  outputSize = [input.shape[2] || 0, input.shape[1] || 0];
  skipped11++;
  const skipTime = (config3.hand.skipTime || 0) > now() - lastTime12;
  const skipFrame = skipped11 < (config3.hand.skipFrames || 0);
  if (config3.skipAllowed && skipTime && skipFrame) {
    return cache4.hands;
  }
  return new Promise(async (resolve) => {
    const skipTimeExtended = 3 * (config3.hand.skipTime || 0) > now() - lastTime12;
    const skipFrameExtended = skipped11 < 3 * (config3.hand.skipFrames || 0);
    if (config3.skipAllowed && cache4.hands.length === config3.hand.maxDetected) {
      cache4.hands = await Promise.all(cache4.boxes.map((handBox) => detectFingers(input, handBox, config3)));
    } else if (config3.skipAllowed && skipTimeExtended && skipFrameExtended && cache4.hands.length > 0) {
      cache4.hands = await Promise.all(cache4.boxes.map((handBox) => detectFingers(input, handBox, config3)));
    } else {
      cache4.boxes = await detectHands(input, config3);
      lastTime12 = now();
      cache4.hands = await Promise.all(cache4.boxes.map((handBox) => detectFingers(input, handBox, config3)));
      skipped11 = 0;
    }
    const oldCache = [...cache4.boxes];
    cache4.boxes.length = 0;
    if (config3.cacheSensitivity > 0) {
      for (let i = 0; i < cache4.hands.length; i++) {
        const boxKpt = square(cache4.hands[i].keypoints, outputSize);
        if (boxKpt.box[2] / (input.shape[2] || 1) > 0.05 && boxKpt.box[3] / (input.shape[1] || 1) > 0.05 && cache4.hands[i].fingerScore && cache4.hands[i].fingerScore > (config3.hand.minConfidence || 0)) {
          const boxScale = scale(boxKpt.box, boxExpandFact);
          const boxScaleRaw = scale(boxKpt.boxRaw, boxExpandFact);
          cache4.boxes.push({ ...oldCache[i], box: boxScale, boxRaw: boxScaleRaw });
        }
      }
    }
    for (let i = 0; i < cache4.hands.length; i++) {
      const bbox = calc(cache4.hands[i].keypoints, outputSize);
      cache4.hands[i].box = bbox.box;
      cache4.hands[i].boxRaw = bbox.boxRaw;
    }
    resolve(cache4.hands);
  });
}

// src/face/liveness.ts
var model13;
var cached2 = [];
var skipped12 = Number.MAX_SAFE_INTEGER;
var lastCount8 = 0;
var lastTime13 = 0;
async function load14(config3) {
  var _a2;
  if (env.initial)
    model13 = null;
  if (!model13)
    model13 = await loadModel((_a2 = config3.face.liveness) == null ? void 0 : _a2.modelPath);
  else if (config3.debug)
    log("cached model:", model13["modelUrl"]);
  return model13;
}
async function predict14(image, config3, idx, count2) {
  var _a2, _b2;
  if (!model13)
    return 0;
  const skipTime = (((_a2 = config3.face.liveness) == null ? void 0 : _a2.skipTime) || 0) > now() - lastTime13;
  const skipFrame = skipped12 < (((_b2 = config3.face.liveness) == null ? void 0 : _b2.skipFrames) || 0);
  if (config3.skipAllowed && skipTime && skipFrame && lastCount8 === count2 && cached2[idx]) {
    skipped12++;
    return cached2[idx];
  }
  skipped12 = 0;
  return new Promise(async (resolve) => {
    const resize = jn.resizeBilinear(image, [(model13 == null ? void 0 : model13.inputs[0].shape) ? model13.inputs[0].shape[2] : 0, (model13 == null ? void 0 : model13.inputs[0].shape) ? model13.inputs[0].shape[1] : 0], false);
    const res = model13 == null ? void 0 : model13.execute(resize);
    const num = (await res.data())[0];
    cached2[idx] = Math.round(100 * num) / 100;
    lastCount8 = count2;
    lastTime13 = now();
    De([resize, res]);
    resolve(cached2[idx]);
  });
}

// src/body/movenetcoords.ts
var movenetcoords_exports = {};
__export(movenetcoords_exports, {
  connected: () => connected3,
  horizontal: () => horizontal,
  kpt: () => kpt3,
  relative: () => relative,
  vertical: () => vertical
});
var kpt3 = [
  "nose",
  "leftEye",
  "rightEye",
  "leftEar",
  "rightEar",
  "leftShoulder",
  "rightShoulder",
  "leftElbow",
  "rightElbow",
  "leftWrist",
  "rightWrist",
  "leftHip",
  "rightHip",
  "leftKnee",
  "rightKnee",
  "leftAnkle",
  "rightAnkle"
];
var horizontal = [
  ["leftEye", "rightEye"],
  ["leftEar", "rightEar"],
  ["leftShoulder", "rightShoulder"],
  ["leftElbow", "rightElbow"],
  ["leftWrist", "rightWrist"],
  ["leftHip", "rightHip"],
  ["leftKnee", "rightKnee"],
  ["leftAnkle", "rightAnkle"]
];
var vertical = [
  ["leftKnee", "leftShoulder"],
  ["rightKnee", "rightShoulder"],
  ["leftAnkle", "leftKnee"],
  ["rightAnkle", "rightKnee"]
];
var relative = [
  [["leftHip", "rightHip"], ["leftShoulder", "rightShoulder"]],
  [["leftElbow", "rightElbow"], ["leftShoulder", "rightShoulder"]]
];
var connected3 = {
  leftLeg: ["leftHip", "leftKnee", "leftAnkle"],
  rightLeg: ["rightHip", "rightKnee", "rightAnkle"],
  torso: ["leftShoulder", "rightShoulder", "rightHip", "leftHip", "leftShoulder"],
  leftArm: ["leftShoulder", "leftElbow", "leftWrist"],
  rightArm: ["rightShoulder", "rightElbow", "rightWrist"],
  head: []
};

// src/body/movenetfix.ts
var maxJitter = 5e-3;
var cache5 = {
  keypoints: [],
  padding: [[0, 0], [0, 0], [0, 0], [0, 0]]
};
function bodyParts(body4) {
  for (const pair of horizontal) {
    const left = body4.keypoints.findIndex((kp2) => kp2.part === pair[0]);
    const right = body4.keypoints.findIndex((kp2) => kp2.part === pair[1]);
    if (body4.keypoints[left] && body4.keypoints[right]) {
      if (body4.keypoints[left].position[0] < body4.keypoints[right].position[0]) {
        const tmp = body4.keypoints[left];
        body4.keypoints[left] = body4.keypoints[right];
        body4.keypoints[right] = tmp;
      }
    }
  }
  for (const pair of vertical) {
    const lower = body4.keypoints.findIndex((kp2) => kp2 && kp2.part === pair[0]);
    const higher = body4.keypoints.findIndex((kp2) => kp2 && kp2.part === pair[1]);
    if (body4.keypoints[lower] && body4.keypoints[higher]) {
      if (body4.keypoints[lower].position[1] < body4.keypoints[higher].position[1]) {
        body4.keypoints.splice(lower, 1);
      }
    }
  }
  for (const [pair, compare2] of relative) {
    const left = body4.keypoints.findIndex((kp2) => kp2 && kp2.part === pair[0]);
    const right = body4.keypoints.findIndex((kp2) => kp2 && kp2.part === pair[1]);
    const leftTo = body4.keypoints.findIndex((kp2) => kp2 && kp2.part === compare2[0]);
    const rightTo = body4.keypoints.findIndex((kp2) => kp2 && kp2.part === compare2[1]);
    if (!body4.keypoints[leftTo] || !body4.keypoints[rightTo])
      continue;
    const distanceLeft = body4.keypoints[left] ? [
      Math.abs(body4.keypoints[leftTo].position[0] - body4.keypoints[left].position[0]),
      Math.abs(body4.keypoints[rightTo].position[0] - body4.keypoints[left].position[0])
    ] : [0, 0];
    const distanceRight = body4.keypoints[right] ? [
      Math.abs(body4.keypoints[rightTo].position[0] - body4.keypoints[right].position[0]),
      Math.abs(body4.keypoints[leftTo].position[0] - body4.keypoints[right].position[0])
    ] : [0, 0];
    if (distanceLeft[0] > distanceLeft[1] || distanceRight[0] > distanceRight[1]) {
      const tmp = body4.keypoints[left];
      body4.keypoints[left] = body4.keypoints[right];
      body4.keypoints[right] = tmp;
    }
  }
}
function jitter(keypoints) {
  for (let i = 0; i < keypoints.length; i++) {
    if (keypoints[i] && cache5.keypoints[i]) {
      const diff = [Math.abs(keypoints[i].positionRaw[0] - cache5.keypoints[i].positionRaw[0]), Math.abs(keypoints[i].positionRaw[1] - cache5.keypoints[i].positionRaw[1])];
      if (diff[0] < maxJitter && diff[1] < maxJitter) {
        keypoints[i] = cache5.keypoints[i];
      } else {
        cache5.keypoints[i] = keypoints[i];
      }
    } else {
      cache5.keypoints[i] = keypoints[i];
    }
  }
  return keypoints;
}
function padInput(input, inputSize10) {
  const t = {};
  if (!input.shape || !input.shape[1] || !input.shape[2])
    return input;
  cache5.padding = [
    [0, 0],
    [input.shape[2] > input.shape[1] ? Math.trunc((input.shape[2] - input.shape[1]) / 2) : 0, input.shape[2] > input.shape[1] ? Math.trunc((input.shape[2] - input.shape[1]) / 2) : 0],
    [input.shape[1] > input.shape[2] ? Math.trunc((input.shape[1] - input.shape[2]) / 2) : 0, input.shape[1] > input.shape[2] ? Math.trunc((input.shape[1] - input.shape[2]) / 2) : 0],
    [0, 0]
  ];
  t.pad = bi(input, cache5.padding);
  t.resize = jn.resizeBilinear(t.pad, [inputSize10, inputSize10]);
  const final = le(t.resize, "int32");
  Object.keys(t).forEach((tensor) => De(t[tensor]));
  return final;
}
function rescaleBody(body4, outputSize2) {
  body4.keypoints = body4.keypoints.filter((kpt4) => kpt4 && kpt4.position);
  for (const kpt4 of body4.keypoints) {
    kpt4.position = [
      kpt4.position[0] * (outputSize2[0] + cache5.padding[2][0] + cache5.padding[2][1]) / outputSize2[0] - cache5.padding[2][0],
      kpt4.position[1] * (outputSize2[1] + cache5.padding[1][0] + cache5.padding[1][1]) / outputSize2[1] - cache5.padding[1][0]
    ];
    kpt4.positionRaw = [
      kpt4.position[0] / outputSize2[0],
      kpt4.position[1] / outputSize2[1]
    ];
  }
  const rescaledBoxes = calc(body4.keypoints.map((pt2) => pt2.position), outputSize2);
  body4.box = rescaledBoxes.box;
  body4.boxRaw = rescaledBoxes.boxRaw;
  return body4;
}

// src/body/movenet.ts
var model14;
var inputSize8 = 0;
var skipped13 = Number.MAX_SAFE_INTEGER;
var cache6 = {
  boxes: [],
  bodies: [],
  last: 0
};
async function load15(config3) {
  if (env.initial)
    model14 = null;
  if (!model14) {
    fakeOps(["size"], config3);
    model14 = await loadModel(config3.body.modelPath);
  } else if (config3.debug)
    log("cached model:", model14["modelUrl"]);
  inputSize8 = model14.inputs[0].shape ? model14.inputs[0].shape[2] : 0;
  if (inputSize8 < 64)
    inputSize8 = 256;
  return model14;
}
async function parseSinglePose(res, config3, image) {
  const kpt4 = res[0][0];
  const keypoints = [];
  let score = 0;
  for (let id2 = 0; id2 < kpt4.length; id2++) {
    score = kpt4[id2][2];
    if (score > config3.body.minConfidence) {
      const positionRaw = [kpt4[id2][1], kpt4[id2][0]];
      keypoints.push({
        score: Math.round(100 * score) / 100,
        part: kpt3[id2],
        positionRaw,
        position: [
          Math.round((image.shape[2] || 0) * positionRaw[0]),
          Math.round((image.shape[1] || 0) * positionRaw[1])
        ]
      });
    }
  }
  score = keypoints.reduce((prev, curr) => curr.score > prev ? curr.score : prev, 0);
  const bodies = [];
  const newBox = calc(keypoints.map((pt2) => pt2.position), [image.shape[2], image.shape[1]]);
  const annotations2 = {};
  for (const [name, indexes] of Object.entries(connected3)) {
    const pt2 = [];
    for (let i = 0; i < indexes.length - 1; i++) {
      const pt0 = keypoints.find((kp2) => kp2.part === indexes[i]);
      const pt1 = keypoints.find((kp2) => kp2.part === indexes[i + 1]);
      if (pt0 && pt1 && pt0.score > (config3.body.minConfidence || 0) && pt1.score > (config3.body.minConfidence || 0))
        pt2.push([pt0.position, pt1.position]);
    }
    annotations2[name] = pt2;
  }
  const body4 = { id: 0, score, box: newBox.box, boxRaw: newBox.boxRaw, keypoints, annotations: annotations2 };
  bodyParts(body4);
  bodies.push(body4);
  return bodies;
}
async function parseMultiPose(res, config3, image) {
  const bodies = [];
  for (let id2 = 0; id2 < res[0].length; id2++) {
    const kpt4 = res[0][id2];
    const totalScore = Math.round(100 * kpt4[51 + 4]) / 100;
    if (totalScore > config3.body.minConfidence) {
      const keypoints = [];
      for (let i = 0; i < 17; i++) {
        const score = kpt4[3 * i + 2];
        if (score > config3.body.minConfidence) {
          const positionRaw = [kpt4[3 * i + 1], kpt4[3 * i + 0]];
          keypoints.push({
            part: kpt3[i],
            score: Math.round(100 * score) / 100,
            positionRaw,
            position: [Math.round((image.shape[2] || 0) * positionRaw[0]), Math.round((image.shape[1] || 0) * positionRaw[1])]
          });
        }
      }
      const newBox = calc(keypoints.map((pt2) => pt2.position), [image.shape[2], image.shape[1]]);
      const annotations2 = {};
      for (const [name, indexes] of Object.entries(connected3)) {
        const pt2 = [];
        for (let i = 0; i < indexes.length - 1; i++) {
          const pt0 = keypoints.find((kp2) => kp2.part === indexes[i]);
          const pt1 = keypoints.find((kp2) => kp2.part === indexes[i + 1]);
          if (pt0 && pt1 && pt0.score > (config3.body.minConfidence || 0) && pt1.score > (config3.body.minConfidence || 0))
            pt2.push([pt0.position, pt1.position]);
        }
        annotations2[name] = pt2;
      }
      const body4 = { id: id2, score: totalScore, box: newBox.box, boxRaw: newBox.boxRaw, keypoints: [...keypoints], annotations: annotations2 };
      bodyParts(body4);
      bodies.push(body4);
    }
  }
  bodies.sort((a, b) => b.score - a.score);
  if (bodies.length > config3.body.maxDetected)
    bodies.length = config3.body.maxDetected;
  return bodies;
}
async function predict15(input, config3) {
  if (!model14 || !(model14 == null ? void 0 : model14.inputs[0].shape))
    return [];
  if (!config3.skipAllowed)
    cache6.boxes.length = 0;
  skipped13++;
  const skipTime = (config3.body.skipTime || 0) > now() - cache6.last;
  const skipFrame = skipped13 < (config3.body.skipFrames || 0);
  if (config3.skipAllowed && skipTime && skipFrame) {
    return cache6.bodies;
  }
  return new Promise(async (resolve) => {
    const t = {};
    skipped13 = 0;
    t.input = padInput(input, inputSize8);
    t.res = model14 == null ? void 0 : model14.execute(t.input);
    cache6.last = now();
    const res = await t.res.array();
    cache6.bodies = t.res.shape[2] === 17 ? await parseSinglePose(res, config3, input) : await parseMultiPose(res, config3, input);
    for (const body4 of cache6.bodies) {
      rescaleBody(body4, [input.shape[2] || 1, input.shape[1] || 1]);
      jitter(body4.keypoints);
    }
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    resolve(cache6.bodies);
  });
}

// src/object/nanodet.ts
var model15;
var last9 = [];
var lastTime14 = 0;
var skipped14 = Number.MAX_SAFE_INTEGER;
var inputSize9 = 0;
var scaleBox = 2.5;
async function load16(config3) {
  if (!model15 || env.initial) {
    model15 = await loadModel(config3.object.modelPath);
    const inputs = Object.values(model15.modelSignature["inputs"]);
    inputSize9 = Array.isArray(inputs) ? parseInt(inputs[0].tensorShape.dim[2].size) : 0;
  } else if (config3.debug)
    log("cached model:", model15["modelUrl"]);
  return model15;
}
async function process4(res, outputShape, config3) {
  let id2 = 0;
  let results = [];
  for (const strideSize of [1, 2, 4]) {
    j(async () => {
      const baseSize = strideSize * 13;
      const scoresT = mr(res.find((a) => a.shape[1] === baseSize ** 2 && (a.shape[2] || 0) === labels.length));
      const featuresT = mr(res.find((a) => a.shape[1] === baseSize ** 2 && (a.shape[2] || 0) < labels.length));
      const boxesMax = featuresT.reshape([-1, 4, featuresT.shape[1] / 4]);
      const boxIdx = await boxesMax.argMax(2).array();
      const scores = await scoresT.array();
      for (let i = 0; i < scoresT.shape[0]; i++) {
        for (let j10 = 0; j10 < scoresT.shape[1]; j10++) {
          const score = scores[i][j10];
          if (score > (config3.object.minConfidence || 0) && j10 !== 61) {
            const cx2 = (0.5 + Math.trunc(i % baseSize)) / baseSize;
            const cy2 = (0.5 + Math.trunc(i / baseSize)) / baseSize;
            const boxOffset = boxIdx[i].map((a) => a * (baseSize / strideSize / inputSize9));
            const [x, y] = [
              cx2 - scaleBox / strideSize * boxOffset[0],
              cy2 - scaleBox / strideSize * boxOffset[1]
            ];
            const [w10, h] = [
              cx2 + scaleBox / strideSize * boxOffset[2] - x,
              cy2 + scaleBox / strideSize * boxOffset[3] - y
            ];
            let boxRaw = [x, y, w10, h];
            boxRaw = boxRaw.map((a) => Math.max(0, Math.min(a, 1)));
            const box = [
              boxRaw[0] * outputShape[0],
              boxRaw[1] * outputShape[1],
              boxRaw[2] * outputShape[0],
              boxRaw[3] * outputShape[1]
            ];
            const result = {
              id: id2++,
              score: Math.round(100 * score) / 100,
              class: j10 + 1,
              label: labels[j10].label,
              box: box.map((a) => Math.trunc(a)),
              boxRaw
            };
            results.push(result);
          }
        }
      }
    });
  }
  res.forEach((t) => De(t));
  const nmsBoxes = results.map((a) => [a.boxRaw[1], a.boxRaw[0], a.boxRaw[3], a.boxRaw[2]]);
  const nmsScores = results.map((a) => a.score);
  let nmsIdx = [];
  if (nmsBoxes && nmsBoxes.length > 0) {
    const nms = await jn.nonMaxSuppressionAsync(nmsBoxes, nmsScores, config3.object.maxDetected, config3.object.iouThreshold, config3.object.minConfidence);
    nmsIdx = await nms.data();
    De(nms);
  }
  results = results.filter((_val, idx) => nmsIdx.includes(idx)).sort((a, b) => b.score - a.score);
  return results;
}
async function predict16(image, config3) {
  const skipTime = (config3.object.skipTime || 0) > now() - lastTime14;
  const skipFrame = skipped14 < (config3.object.skipFrames || 0);
  if (config3.skipAllowed && skipTime && skipFrame && last9.length > 0) {
    skipped14++;
    return last9;
  }
  skipped14 = 0;
  if (!env.kernels.includes("mod") || !env.kernels.includes("sparsetodense"))
    return last9;
  return new Promise(async (resolve) => {
    const outputSize2 = [image.shape[2] || 0, image.shape[1] || 0];
    const resize = jn.resizeBilinear(image, [inputSize9, inputSize9], false);
    const norm = xe(resize, constants.tf255);
    const transpose = norm.transpose([0, 3, 1, 2]);
    De(norm);
    De(resize);
    let objectT;
    if (config3.object.enabled)
      objectT = model15.execute(transpose);
    lastTime14 = now();
    De(transpose);
    const obj = await process4(objectT, outputSize2, config3);
    last9 = obj;
    resolve(obj);
  });
}

// src/body/posenetutils.ts
var partNames = [
  "nose",
  "leftEye",
  "rightEye",
  "leftEar",
  "rightEar",
  "leftShoulder",
  "rightShoulder",
  "leftElbow",
  "rightElbow",
  "leftWrist",
  "rightWrist",
  "leftHip",
  "rightHip",
  "leftKnee",
  "rightKnee",
  "leftAnkle",
  "rightAnkle"
];
var count = partNames.length;
var partIds = partNames.reduce((result, jointName, i) => {
  result[jointName] = i;
  return result;
}, {});
var connectedPartNames = [
  ["leftHip", "leftShoulder"],
  ["leftElbow", "leftShoulder"],
  ["leftElbow", "leftWrist"],
  ["leftHip", "leftKnee"],
  ["leftKnee", "leftAnkle"],
  ["rightHip", "rightShoulder"],
  ["rightElbow", "rightShoulder"],
  ["rightElbow", "rightWrist"],
  ["rightHip", "rightKnee"],
  ["rightKnee", "rightAnkle"],
  ["leftShoulder", "rightShoulder"],
  ["leftHip", "rightHip"]
];
var connectedPartIndices = connectedPartNames.map(([jointNameA, jointNameB]) => [partIds[jointNameA], partIds[jointNameB]]);
var poseChain = [
  ["nose", "leftEye"],
  ["leftEye", "leftEar"],
  ["nose", "rightEye"],
  ["rightEye", "rightEar"],
  ["nose", "leftShoulder"],
  ["leftShoulder", "leftElbow"],
  ["leftElbow", "leftWrist"],
  ["leftShoulder", "leftHip"],
  ["leftHip", "leftKnee"],
  ["leftKnee", "leftAnkle"],
  ["nose", "rightShoulder"],
  ["rightShoulder", "rightElbow"],
  ["rightElbow", "rightWrist"],
  ["rightShoulder", "rightHip"],
  ["rightHip", "rightKnee"],
  ["rightKnee", "rightAnkle"]
];
function getBoundingBox(keypoints) {
  const coord = keypoints.reduce(({ maxX, maxY, minX, minY }, { position: { x, y } }) => ({
    maxX: Math.max(maxX, x),
    maxY: Math.max(maxY, y),
    minX: Math.min(minX, x),
    minY: Math.min(minY, y)
  }), {
    maxX: Number.NEGATIVE_INFINITY,
    maxY: Number.NEGATIVE_INFINITY,
    minX: Number.POSITIVE_INFINITY,
    minY: Number.POSITIVE_INFINITY
  });
  return [coord.minX, coord.minY, coord.maxX - coord.minX, coord.maxY - coord.minY];
}
function scalePoses(poses, [height, width], [inputResolutionHeight, inputResolutionWidth]) {
  const scaleY = height / inputResolutionHeight;
  const scaleX = width / inputResolutionWidth;
  const scalePose = (pose, i) => ({
    id: i,
    score: pose.score,
    boxRaw: [pose.box[0] / inputResolutionWidth, pose.box[1] / inputResolutionHeight, pose.box[2] / inputResolutionWidth, pose.box[3] / inputResolutionHeight],
    box: [Math.trunc(pose.box[0] * scaleX), Math.trunc(pose.box[1] * scaleY), Math.trunc(pose.box[2] * scaleX), Math.trunc(pose.box[3] * scaleY)],
    keypoints: pose.keypoints.map(({ score, part, position }) => ({
      score,
      part,
      position: [Math.trunc(position.x * scaleX), Math.trunc(position.y * scaleY)],
      positionRaw: [position.x / inputResolutionHeight, position.y / inputResolutionHeight]
    })),
    annotations: {}
  });
  const scaledPoses = poses.map((pose, i) => scalePose(pose, i));
  return scaledPoses;
}
var MaxHeap = class {
  constructor(maxSize2, getElementValue) {
    __publicField(this, "priorityQueue");
    __publicField(this, "numberOfElements");
    __publicField(this, "getElementValue");
    this.priorityQueue = new Array(maxSize2);
    this.numberOfElements = -1;
    this.getElementValue = getElementValue;
  }
  enqueue(x) {
    this.priorityQueue[++this.numberOfElements] = x;
    this.swim(this.numberOfElements);
  }
  dequeue() {
    const max = this.priorityQueue[0];
    this.exchange(0, this.numberOfElements--);
    this.sink(0);
    this.priorityQueue[this.numberOfElements + 1] = null;
    return max;
  }
  empty() {
    return this.numberOfElements === -1;
  }
  size() {
    return this.numberOfElements + 1;
  }
  all() {
    return this.priorityQueue.slice(0, this.numberOfElements + 1);
  }
  max() {
    return this.priorityQueue[0];
  }
  swim(k) {
    while (k > 0 && this.less(Math.floor(k / 2), k)) {
      this.exchange(k, Math.floor(k / 2));
      k = Math.floor(k / 2);
    }
  }
  sink(k) {
    while (2 * k <= this.numberOfElements) {
      let j10 = 2 * k;
      if (j10 < this.numberOfElements && this.less(j10, j10 + 1))
        j10++;
      if (!this.less(k, j10))
        break;
      this.exchange(k, j10);
      k = j10;
    }
  }
  getValueAt(i) {
    return this.getElementValue(this.priorityQueue[i]);
  }
  less(i, j10) {
    return this.getValueAt(i) < this.getValueAt(j10);
  }
  exchange(i, j10) {
    const t = this.priorityQueue[i];
    this.priorityQueue[i] = this.priorityQueue[j10];
    this.priorityQueue[j10] = t;
  }
};
function getOffsetPoint(y, x, keypoint, offsets) {
  return {
    y: offsets.get(y, x, keypoint),
    x: offsets.get(y, x, keypoint + count)
  };
}
function getImageCoords(part, outputStride2, offsets) {
  const { heatmapY, heatmapX, id: keypoint } = part;
  const { y, x } = getOffsetPoint(heatmapY, heatmapX, keypoint, offsets);
  return {
    x: part.heatmapX * outputStride2 + x,
    y: part.heatmapY * outputStride2 + y
  };
}
function clamp(a, min, max) {
  if (a < min)
    return min;
  if (a > max)
    return max;
  return a;
}
function squaredDistance(y12, x12, y22, x22) {
  const dy2 = y22 - y12;
  const dx2 = x22 - x12;
  return dy2 * dy2 + dx2 * dx2;
}
function addVectors(a, b) {
  return { x: a.x + b.x, y: a.y + b.y };
}

// src/body/posenet.ts
var model16;
var poseNetOutputs = ["MobilenetV1/offset_2/BiasAdd", "MobilenetV1/heatmap_2/BiasAdd", "MobilenetV1/displacement_fwd_2/BiasAdd", "MobilenetV1/displacement_bwd_2/BiasAdd"];
var localMaximumRadius = 1;
var outputStride = 16;
var squaredNmsRadius = 50 ** 2;
function traverse(edgeId, sourceKeypoint, targetId, scores, offsets, displacements, offsetRefineStep = 2) {
  const getDisplacement = (point2) => ({
    y: displacements.get(point2.y, point2.x, edgeId),
    x: displacements.get(point2.y, point2.x, displacements.shape[2] / 2 + edgeId)
  });
  const getStridedIndexNearPoint = (point2, height2, width2) => ({
    y: clamp(Math.round(point2.y / outputStride), 0, height2 - 1),
    x: clamp(Math.round(point2.x / outputStride), 0, width2 - 1)
  });
  const [height, width] = scores.shape;
  const sourceKeypointIndices = getStridedIndexNearPoint(sourceKeypoint.position, height, width);
  const displacement = getDisplacement(sourceKeypointIndices);
  const displacedPoint = addVectors(sourceKeypoint.position, displacement);
  let targetKeypoint = displacedPoint;
  for (let i = 0; i < offsetRefineStep; i++) {
    const targetKeypointIndices = getStridedIndexNearPoint(targetKeypoint, height, width);
    const offsetPoint = getOffsetPoint(targetKeypointIndices.y, targetKeypointIndices.x, targetId, offsets);
    targetKeypoint = addVectors({ x: targetKeypointIndices.x * outputStride, y: targetKeypointIndices.y * outputStride }, { x: offsetPoint.x, y: offsetPoint.y });
  }
  const targetKeyPointIndices = getStridedIndexNearPoint(targetKeypoint, height, width);
  const score = scores.get(targetKeyPointIndices.y, targetKeyPointIndices.x, targetId);
  return { position: targetKeypoint, part: partNames[targetId], score };
}
function decodePose(root, scores, offsets, displacementsFwd, displacementsBwd) {
  const tuples = poseChain.map(([parentJoinName, childJoinName]) => [partIds[parentJoinName], partIds[childJoinName]]);
  const edgesFwd = tuples.map(([, childJointId]) => childJointId);
  const edgesBwd = tuples.map(([parentJointId]) => parentJointId);
  const numParts = scores.shape[2];
  const numEdges = edgesFwd.length;
  const keypoints = new Array(numParts);
  const rootPoint = getImageCoords(root.part, outputStride, offsets);
  keypoints[root.part.id] = {
    score: root.score,
    part: partNames[root.part.id],
    position: rootPoint
  };
  for (let edge = numEdges - 1; edge >= 0; --edge) {
    const sourceId = edgesFwd[edge];
    const targetId = edgesBwd[edge];
    if (keypoints[sourceId] && !keypoints[targetId]) {
      keypoints[targetId] = traverse(edge, keypoints[sourceId], targetId, scores, offsets, displacementsBwd);
    }
  }
  for (let edge = 0; edge < numEdges; ++edge) {
    const sourceId = edgesBwd[edge];
    const targetId = edgesFwd[edge];
    if (keypoints[sourceId] && !keypoints[targetId]) {
      keypoints[targetId] = traverse(edge, keypoints[sourceId], targetId, scores, offsets, displacementsFwd);
    }
  }
  return keypoints;
}
function scoreIsMaximumInLocalWindow(keypointId, score, heatmapY, heatmapX, scores) {
  const [height, width] = scores.shape;
  let localMaximum = true;
  const yStart = Math.max(heatmapY - localMaximumRadius, 0);
  const yEnd = Math.min(heatmapY + localMaximumRadius + 1, height);
  for (let yCurrent = yStart; yCurrent < yEnd; ++yCurrent) {
    const xStart = Math.max(heatmapX - localMaximumRadius, 0);
    const xEnd = Math.min(heatmapX + localMaximumRadius + 1, width);
    for (let xCurrent = xStart; xCurrent < xEnd; ++xCurrent) {
      if (scores.get(yCurrent, xCurrent, keypointId) > score) {
        localMaximum = false;
        break;
      }
    }
    if (!localMaximum)
      break;
  }
  return localMaximum;
}
function buildPartWithScoreQueue(minConfidence2, scores) {
  const [height, width, numKeypoints] = scores.shape;
  const queue = new MaxHeap(height * width * numKeypoints, ({ score }) => score);
  for (let heatmapY = 0; heatmapY < height; ++heatmapY) {
    for (let heatmapX = 0; heatmapX < width; ++heatmapX) {
      for (let keypointId = 0; keypointId < numKeypoints; ++keypointId) {
        const score = scores.get(heatmapY, heatmapX, keypointId);
        if (score < minConfidence2)
          continue;
        if (scoreIsMaximumInLocalWindow(keypointId, score, heatmapY, heatmapX, scores))
          queue.enqueue({ score, part: { heatmapY, heatmapX, id: keypointId } });
      }
    }
  }
  return queue;
}
function withinRadius(poses, { x, y }, keypointId) {
  return poses.some(({ keypoints }) => {
    var _a2;
    const correspondingKeypoint = (_a2 = keypoints[keypointId]) == null ? void 0 : _a2.position;
    if (!correspondingKeypoint)
      return false;
    return squaredDistance(y, x, correspondingKeypoint.y, correspondingKeypoint.x) <= squaredNmsRadius;
  });
}
function getInstanceScore(existingPoses, keypoints) {
  const notOverlappedKeypointScores = keypoints.reduce((result, { position, score }, keypointId) => {
    if (!withinRadius(existingPoses, position, keypointId))
      result += score;
    return result;
  }, 0);
  return notOverlappedKeypointScores / keypoints.length;
}
function decode(offsets, scores, displacementsFwd, displacementsBwd, maxDetected, minConfidence2) {
  const poses = [];
  const queue = buildPartWithScoreQueue(minConfidence2, scores);
  while (poses.length < maxDetected && !queue.empty()) {
    const root = queue.dequeue();
    const rootImageCoords = getImageCoords(root.part, outputStride, offsets);
    if (withinRadius(poses, rootImageCoords, root.part.id))
      continue;
    let keypoints = decodePose(root, scores, offsets, displacementsFwd, displacementsBwd);
    keypoints = keypoints.filter((a) => a.score > minConfidence2);
    const score = getInstanceScore(poses, keypoints);
    const box = getBoundingBox(keypoints);
    if (score > minConfidence2)
      poses.push({ keypoints, box, score: Math.round(100 * score) / 100 });
  }
  return poses;
}
async function predict17(input, config3) {
  const res = j(() => {
    if (!model16.inputs[0].shape)
      return [];
    const resized = jn.resizeBilinear(input, [model16.inputs[0].shape[2], model16.inputs[0].shape[1]]);
    const normalized = ge(xe(le(resized, "float32"), 127.5), 1);
    const results = model16.execute(normalized, poseNetOutputs);
    const results3d = results.map((y) => mr(y, [0]));
    results3d[1] = Hs(results3d[1]);
    return results3d;
  });
  const buffers = await Promise.all(res.map((tensor) => tensor.buffer()));
  for (const t of res)
    De(t);
  const decoded = await decode(buffers[0], buffers[1], buffers[2], buffers[3], config3.body.maxDetected, config3.body.minConfidence);
  if (!model16.inputs[0].shape)
    return [];
  const scaled = scalePoses(decoded, [input.shape[1], input.shape[2]], [model16.inputs[0].shape[2], model16.inputs[0].shape[1]]);
  return scaled;
}
async function load17(config3) {
  if (!model16 || env.initial)
    model16 = await loadModel(config3.body.modelPath);
  else if (config3.debug)
    log("cached model:", model16["modelUrl"]);
  return model16;
}

// src/segmentation/segmentation.ts
var model17;
var busy = false;
async function load18(config3) {
  if (!model17 || env.initial)
    model17 = await loadModel(config3.segmentation.modelPath);
  else if (config3.debug)
    log("cached model:", model17["modelUrl"]);
  return model17;
}
async function process5(input, background, config3) {
  var _a2, _b2;
  if (busy)
    return { data: [], canvas: null, alpha: null };
  busy = true;
  if (!model17)
    await load18(config3);
  const inputImage = await process2(input, config3);
  const width = ((_a2 = inputImage.tensor) == null ? void 0 : _a2.shape[2]) || 0;
  const height = ((_b2 = inputImage.tensor) == null ? void 0 : _b2.shape[1]) || 0;
  if (!inputImage.tensor)
    return { data: [], canvas: null, alpha: null };
  const t = {};
  t.resize = jn.resizeBilinear(inputImage.tensor, [model17.inputs[0].shape ? model17.inputs[0].shape[1] : 0, model17.inputs[0].shape ? model17.inputs[0].shape[2] : 0], false);
  De(inputImage.tensor);
  t.norm = xe(t.resize, constants.tf255);
  t.res = model17.execute(t.norm);
  t.squeeze = mr(t.res, 0);
  if (t.squeeze.shape[2] === 2) {
    t.softmax = gb(t.squeeze);
    [t.bg, t.fg] = Fs(t.softmax, 2);
    t.expand = Pn(t.fg, 2);
    t.pad = Pn(t.expand, 0);
    t.crop = jn.cropAndResize(t.pad, [[0, 0, 0.5, 0.5]], [0], [width, height]);
    t.data = mr(t.crop, 0);
  } else {
    t.data = jn.resizeBilinear(t.squeeze, [height, width]);
  }
  const data = Array.from(await t.data.data());
  if (env.node && !env.Canvas && typeof ImageData === "undefined") {
    if (config3.debug)
      log("canvas support missing");
    Object.keys(t).forEach((tensor) => De(t[tensor]));
    return { data, canvas: null, alpha: null };
  }
  const alphaCanvas = canvas(width, height);
  if (Pk)
    await Pk.toPixels(t.data, alphaCanvas);
  const alphaCtx = alphaCanvas.getContext("2d");
  if (config3.segmentation.blur && config3.segmentation.blur > 0)
    alphaCtx.filter = `blur(${config3.segmentation.blur}px)`;
  const alphaData = alphaCtx.getImageData(0, 0, width, height);
  const compositeCanvas = canvas(width, height);
  const compositeCtx = compositeCanvas.getContext("2d");
  if (inputImage.canvas)
    compositeCtx.drawImage(inputImage.canvas, 0, 0);
  compositeCtx.globalCompositeOperation = "darken";
  if (config3.segmentation.blur && config3.segmentation.blur > 0)
    compositeCtx.filter = `blur(${config3.segmentation.blur}px)`;
  compositeCtx.drawImage(alphaCanvas, 0, 0);
  compositeCtx.globalCompositeOperation = "source-over";
  compositeCtx.filter = "none";
  const compositeData = compositeCtx.getImageData(0, 0, width, height);
  for (let i = 0; i < width * height; i++)
    compositeData.data[4 * i + 3] = alphaData.data[4 * i + 0];
  compositeCtx.putImageData(compositeData, 0, 0);
  let mergedCanvas = null;
  if (background && compositeCanvas) {
    mergedCanvas = canvas(width, height);
    const bgImage = await process2(background, config3);
    De(bgImage.tensor);
    const ctxMerge = mergedCanvas.getContext("2d");
    ctxMerge.drawImage(bgImage.canvas, 0, 0, mergedCanvas.width, mergedCanvas.height);
    ctxMerge.drawImage(compositeCanvas, 0, 0);
  }
  Object.keys(t).forEach((tensor) => De(t[tensor]));
  busy = false;
  return { data, canvas: compositeCanvas, alpha: alphaCanvas };
}

// src/models.ts
var Models = class {
  constructor() {
    __publicField(this, "ssrnetage", null);
    __publicField(this, "gear", null);
    __publicField(this, "blazeposedetect", null);
    __publicField(this, "blazepose", null);
    __publicField(this, "centernet", null);
    __publicField(this, "efficientpose", null);
    __publicField(this, "mobilefacenet", null);
    __publicField(this, "emotion", null);
    __publicField(this, "facedetect", null);
    __publicField(this, "faceiris", null);
    __publicField(this, "facemesh", null);
    __publicField(this, "faceres", null);
    __publicField(this, "ssrnetgender", null);
    __publicField(this, "handpose", null);
    __publicField(this, "handskeleton", null);
    __publicField(this, "handtrack", null);
    __publicField(this, "liveness", null);
    __publicField(this, "movenet", null);
    __publicField(this, "nanodet", null);
    __publicField(this, "posenet", null);
    __publicField(this, "segmentation", null);
    __publicField(this, "antispoof", null);
  }
};
function reset(instance) {
  for (const model18 of Object.keys(instance.models))
    instance.models[model18] = null;
}
async function load19(instance) {
  var _a2, _b2, _c, _d2, _e2, _f, _g2, _h, _i, _j2, _k2, _l2, _m2, _n2, _o2, _p2, _q2, _r2, _s2, _t2, _u2, _v2, _w2, _x2, _y2, _z2, _A2, _B2, _C2, _D2, _E2;
  if (env.initial)
    reset(instance);
  if (instance.config.hand.enabled) {
    if (!instance.models.handpose && ((_b2 = (_a2 = instance.config.hand.detector) == null ? void 0 : _a2.modelPath) == null ? void 0 : _b2.includes("handdetect")))
      [instance.models.handpose, instance.models.handskeleton] = await load13(instance.config);
    if (!instance.models.handskeleton && instance.config.hand.landmarks && ((_d2 = (_c = instance.config.hand.detector) == null ? void 0 : _c.modelPath) == null ? void 0 : _d2.includes("handdetect")))
      [instance.models.handpose, instance.models.handskeleton] = await load13(instance.config);
  }
  if (instance.config.body.enabled && !instance.models.blazepose && ((_f = (_e2 = instance.config.body) == null ? void 0 : _e2.modelPath) == null ? void 0 : _f.includes("blazepose")))
    instance.models.blazepose = loadPose(instance.config);
  if (instance.config.body.enabled && !instance.models.blazeposedetect && instance.config.body["detector"] && instance.config.body["detector"]["modelPath"])
    instance.models.blazeposedetect = loadDetect(instance.config);
  if (instance.config.body.enabled && !instance.models.efficientpose && ((_h = (_g2 = instance.config.body) == null ? void 0 : _g2.modelPath) == null ? void 0 : _h.includes("efficientpose")))
    instance.models.efficientpose = load7(instance.config);
  if (instance.config.body.enabled && !instance.models.movenet && ((_j2 = (_i = instance.config.body) == null ? void 0 : _i.modelPath) == null ? void 0 : _j2.includes("movenet")))
    instance.models.movenet = load15(instance.config);
  if (instance.config.body.enabled && !instance.models.posenet && ((_l2 = (_k2 = instance.config.body) == null ? void 0 : _k2.modelPath) == null ? void 0 : _l2.includes("posenet")))
    instance.models.posenet = load17(instance.config);
  if (instance.config.face.enabled && !instance.models.facedetect)
    instance.models.facedetect = load5(instance.config);
  if (instance.config.face.enabled && ((_m2 = instance.config.face.antispoof) == null ? void 0 : _m2.enabled) && !instance.models.antispoof)
    instance.models.antispoof = load4(instance.config);
  if (instance.config.face.enabled && ((_n2 = instance.config.face.liveness) == null ? void 0 : _n2.enabled) && !instance.models.liveness)
    instance.models.liveness = load14(instance.config);
  if (instance.config.face.enabled && ((_o2 = instance.config.face.description) == null ? void 0 : _o2.enabled) && !instance.models.faceres)
    instance.models.faceres = load12(instance.config);
  if (instance.config.face.enabled && ((_p2 = instance.config.face.emotion) == null ? void 0 : _p2.enabled) && !instance.models.emotion)
    instance.models.emotion = load8(instance.config);
  if (instance.config.face.enabled && ((_q2 = instance.config.face.iris) == null ? void 0 : _q2.enabled) && !((_r2 = instance.config.face.attention) == null ? void 0 : _r2.enabled) && !instance.models.faceiris)
    instance.models.faceiris = load10(instance.config);
  if (instance.config.face.enabled && ((_s2 = instance.config.face.mesh) == null ? void 0 : _s2.enabled) && !instance.models.facemesh)
    instance.models.facemesh = load11(instance.config);
  if (instance.config.face.enabled && ((_t2 = instance.config.face["gear"]) == null ? void 0 : _t2.enabled) && !instance.models.gear)
    instance.models.gear = load(instance.config);
  if (instance.config.face.enabled && ((_u2 = instance.config.face["ssrnet"]) == null ? void 0 : _u2.enabled) && !instance.models.ssrnetage)
    instance.models.ssrnetage = load2(instance.config);
  if (instance.config.face.enabled && ((_v2 = instance.config.face["ssrnet"]) == null ? void 0 : _v2.enabled) && !instance.models.ssrnetgender)
    instance.models.ssrnetgender = load3(instance.config);
  if (instance.config.face.enabled && ((_w2 = instance.config.face["mobilefacenet"]) == null ? void 0 : _w2.enabled) && !instance.models.mobilefacenet)
    instance.models.mobilefacenet = load9(instance.config);
  if (instance.config.hand.enabled && !instance.models.handtrack && ((_y2 = (_x2 = instance.config.hand.detector) == null ? void 0 : _x2.modelPath) == null ? void 0 : _y2.includes("handtrack")))
    instance.models.handtrack = loadDetect2(instance.config);
  if (instance.config.hand.enabled && instance.config.hand.landmarks && !instance.models.handskeleton && ((_A2 = (_z2 = instance.config.hand.detector) == null ? void 0 : _z2.modelPath) == null ? void 0 : _A2.includes("handtrack")))
    instance.models.handskeleton = loadSkeleton(instance.config);
  if (instance.config.object.enabled && !instance.models.centernet && ((_C2 = (_B2 = instance.config.object) == null ? void 0 : _B2.modelPath) == null ? void 0 : _C2.includes("centernet")))
    instance.models.centernet = load6(instance.config);
  if (instance.config.object.enabled && !instance.models.nanodet && ((_E2 = (_D2 = instance.config.object) == null ? void 0 : _D2.modelPath) == null ? void 0 : _E2.includes("nanodet")))
    instance.models.nanodet = load16(instance.config);
  if (instance.config.segmentation.enabled && !instance.models.segmentation)
    instance.models.segmentation = load18(instance.config);
  for await (const model18 of Object.keys(instance.models)) {
    if (instance.models[model18] && typeof instance.models[model18] !== "undefined")
      instance.models[model18] = await instance.models[model18];
  }
}
async function validate2(instance) {
  const simpleOps = ["const", "placeholder", "noop", "pad", "squeeze", "add", "sub", "mul", "div"];
  for (const defined of Object.keys(instance.models)) {
    const model18 = instance.models[defined];
    if (!model18)
      continue;
    const ops = [];
    const executor = model18 == null ? void 0 : model18.executor;
    if (executor && executor.graph.nodes) {
      for (const kernel of Object.values(executor.graph.nodes)) {
        const op2 = kernel.op.toLowerCase();
        if (!ops.includes(op2))
          ops.push(op2);
      }
    } else {
      if (!executor && instance.config.debug)
        log("model signature not determined:", defined);
    }
    const missing = [];
    for (const op2 of ops) {
      if (!simpleOps.includes(op2) && !instance.env.kernels.includes(op2) && !instance.env.kernels.includes(op2.replace("_", "")) && !instance.env.kernels.includes(op2.replace("native", "")) && !instance.env.kernels.includes(op2.replace("v2", ""))) {
        missing.push(op2);
      }
    }
    if (instance.config.debug && missing.length > 0)
      log("model validation failed:", defined, missing);
  }
}

// src/tfjs/humangl.ts
var config2 = {
  name: "humangl",
  priority: 999,
  canvas: null,
  gl: null,
  extensions: [],
  webGLattr: {
    alpha: false,
    antialias: false,
    premultipliedAlpha: false,
    preserveDrawingBuffer: false,
    depth: false,
    stencil: false,
    failIfMajorPerformanceCaveat: false,
    desynchronized: true
  }
};
function extensions() {
  const gl2 = config2.gl;
  if (!gl2)
    return;
  config2.extensions = gl2.getSupportedExtensions();
}
async function register(instance) {
  var _a2;
  if (instance.config.backend !== "humangl")
    return;
  if (config2.name in ds().registry && (!config2.gl || !config2.gl.getParameter(config2.gl.VERSION))) {
    log("error: humangl backend invalid context");
    reset(instance);
  }
  if (!wpe(config2.name)) {
    try {
      config2.canvas = await canvas(100, 100);
    } catch (err) {
      log("error: cannot create canvas:", err);
      return;
    }
    try {
      config2.gl = (_a2 = config2.canvas) == null ? void 0 : _a2.getContext("webgl2", config2.webGLattr);
      const glv2 = config2.gl.getParameter(config2.gl.VERSION).includes("2.0");
      if (!glv2) {
        log("override: using fallback webgl backend as webgl 2.0 is not detected");
        instance.config.backend = "webgl";
        return;
      }
      if (config2.canvas) {
        config2.canvas.addEventListener("webglcontextlost", async (e) => {
          log("error: humangl:", e.type);
          log("possible browser memory leak using webgl or conflict with multiple backend registrations");
          instance.emit("error");
          throw new Error("backend error: webgl context lost");
        });
        config2.canvas.addEventListener("webglcontextrestored", (e) => {
          log("error: humangl context restored:", e);
        });
        config2.canvas.addEventListener("webglcontextcreationerror", (e) => {
          log("error: humangl context create:", e);
        });
      }
    } catch (err) {
      log("error: cannot get WebGL context:", err);
      return;
    }
    try {
      q5(2, config2.gl);
    } catch (err) {
      log("error: cannot set WebGL context:", err);
      return;
    }
    try {
      const ctx = new em(config2.gl);
      vp(config2.name, () => new K1(ctx), config2.priority);
    } catch (err) {
      log("error: cannot register WebGL backend:", err);
      return;
    }
    try {
      const kernels = am("webgl");
      kernels.forEach((kernelConfig) => {
        const newKernelConfig = { ...kernelConfig, backendName: config2.name };
        Fl(newKernelConfig);
      });
    } catch (err) {
      log("error: cannot update WebGL backend registration:", err);
      return;
    }
    const current = HA().getGPGPUContext ? HA().getGPGPUContext().gl : null;
    if (current) {
      log(`humangl webgl version:${current.getParameter(current.VERSION)} renderer:${current.getParameter(current.RENDERER)}`);
    } else {
      log("error: no current gl context:", current, config2.gl);
      return;
    }
    try {
      lk.set("WEBGL_VERSION", 2);
    } catch (err) {
      log("error: cannot set WebGL backend flags:", err);
      return;
    }
    extensions();
    log("backend registered:", config2.name);
  }
}

// src/tfjs/backend.ts
function registerCustomOps() {
  if (!env.kernels.includes("mod")) {
    const kernelMod = {
      kernelName: "Mod",
      backendName: vpe(),
      kernelFunc: (op2) => j(() => ge(op2.inputs.a, V(xe(op2.inputs.a, op2.inputs.b), op2.inputs.b)))
    };
    Fl(kernelMod);
    env.kernels.push("mod");
  }
  if (!env.kernels.includes("floormod")) {
    const kernelMod = {
      kernelName: "FloorMod",
      backendName: vpe(),
      kernelFunc: (op2) => j(() => tS(op2.inputs.a / op2.inputs.b) * op2.inputs.b + zD(op2.inputs.a, op2.inputs.b))
    };
    Fl(kernelMod);
    env.kernels.push("floormod");
  }
}
async function check(instance, force = false) {
  instance.state = "backend";
  if (force || env.initial || instance.config.backend && instance.config.backend.length > 0 && vpe() !== instance.config.backend) {
    const timeStamp = now();
    if (instance.config.backend && instance.config.backend.length > 0) {
      if (typeof window === "undefined" && typeof WorkerGlobalScope !== "undefined" && instance.config.debug) {
        if (instance.config.debug)
          log("running inside web worker");
      }
      if (env.browser && instance.config.backend === "tensorflow") {
        if (instance.config.debug)
          log("override: backend set to tensorflow while running in browser");
        instance.config.backend = "humangl";
      }
      if (env.node && (instance.config.backend === "webgl" || instance.config.backend === "humangl")) {
        if (instance.config.debug)
          log(`override: backend set to ${instance.config.backend} while running in nodejs`);
        instance.config.backend = "tensorflow";
      }
      if (env.browser && instance.config.backend === "webgpu") {
        if (typeof navigator === "undefined" || typeof navigator["gpu"] === "undefined") {
          log("override: backend set to webgpu but browser does not support webgpu");
          instance.config.backend = "humangl";
        } else {
          const adapter = await navigator["gpu"].requestAdapter();
          if (instance.config.debug)
            log("enumerated webgpu adapter:", adapter);
        }
      }
      if (instance.config.backend === "humangl")
        await register(instance);
      const available = Object.keys(ds().registryFactory);
      if (instance.config.debug)
        log("available backends:", available);
      if (!available.includes(instance.config.backend)) {
        log(`error: backend ${instance.config.backend} not found in registry`);
        instance.config.backend = env.node ? "tensorflow" : "webgl";
        if (instance.config.debug)
          log(`override: setting backend ${instance.config.backend}`);
      }
      if (instance.config.debug)
        log("setting backend:", instance.config.backend);
      if (instance.config.backend === "wasm") {
        if (instance.config.debug)
          log("wasm path:", instance.config.wasmPath);
        if (typeof (tfjs_esm_exports == null ? void 0 : tfjs_esm_exports.setWasmPaths) !== "undefined")
          await yhe(instance.config.wasmPath, instance.config.wasmPlatformFetch);
        else
          throw new Error("backend error: attempting to use wasm backend but wasm path is not set");
        const simd = await K().getAsync("WASM_HAS_SIMD_SUPPORT");
        const mt2 = await K().getAsync("WASM_HAS_MULTITHREAD_SUPPORT");
        if (instance.config.debug)
          log(`wasm execution: ${simd ? "SIMD" : "no SIMD"} ${mt2 ? "multithreaded" : "singlethreaded"}`);
        if (instance.config.debug && !simd)
          log("warning: wasm simd support is not enabled");
      }
      try {
        await bpe(instance.config.backend);
        await ype();
        init();
      } catch (err) {
        log("error: cannot set backend:", instance.config.backend, err);
        return false;
      }
    }
    if (vpe() === "humangl") {
      lk.set("CHECK_COMPUTATION_FOR_ERRORS", false);
      lk.set("WEBGL_CPU_FORWARD", true);
      lk.set("WEBGL_USE_SHAPES_UNIFORMS", true);
      lk.set("CPU_HANDOFF_SIZE_THRESHOLD", 256);
      if (typeof instance.config["deallocate"] !== "undefined" && instance.config["deallocate"]) {
        log("changing webgl: WEBGL_DELETE_TEXTURE_THRESHOLD:", true);
        lk.set("WEBGL_DELETE_TEXTURE_THRESHOLD", 0);
      }
      if (HA().getGPGPUContext) {
        const gl2 = await HA().getGPGPUContext().gl;
        if (instance.config.debug)
          log(`gl version:${gl2.getParameter(gl2.VERSION)} renderer:${gl2.getParameter(gl2.RENDERER)}`);
      }
    }
    if (vpe() === "webgpu") {
    }
    dpe();
    await ype();
    instance.performance.initBackend = Math.trunc(now() - timeStamp);
    instance.config.backend = vpe();
    await env.updateBackend();
    registerCustomOps();
  }
  return true;
}
function fakeOps(kernelNames, config3) {
  for (const kernelName of kernelNames) {
    const kernelConfig = {
      kernelName,
      backendName: config3.backend,
      kernelFunc: () => {
        if (config3.debug)
          log("kernelFunc", kernelName, config3.backend);
      }
    };
    Fl(kernelConfig);
  }
  env.kernels = am(vpe()).map((kernel) => kernel.kernelName.toLowerCase());
}

// src/draw/draw.ts
var draw_exports = {};
__export(draw_exports, {
  all: () => all,
  body: () => body,
  canvas: () => canvas2,
  face: () => face,
  gesture: () => gesture,
  hand: () => hand,
  object: () => object,
  options: () => options3,
  person: () => person
});

// src/draw/primitives.ts
var getCanvasContext = (input) => {
  if (!input)
    log("draw error: invalid canvas");
  else if (!input.getContext)
    log("draw error: canvas context not defined");
  else {
    const ctx = input.getContext("2d");
    if (!ctx)
      log("draw error: cannot get canvas context");
    else
      return ctx;
  }
  return null;
};
var rad2deg = (theta) => Math.round(theta * 180 / Math.PI);
var colorDepth = (z, opt2) => {
  if (!opt2.useDepth || typeof z === "undefined")
    return opt2.color;
  const rgb2 = Uint8ClampedArray.from([127 + 2 * z, 127 - 2 * z, 255]);
  const color = `rgba(${rgb2[0]}, ${rgb2[1]}, ${rgb2[2]}, ${opt2.alpha})`;
  return color;
};
function point(ctx, x, y, z, localOptions) {
  ctx.fillStyle = colorDepth(z, localOptions);
  ctx.beginPath();
  ctx.arc(x, y, localOptions.pointSize, 0, 2 * Math.PI);
  ctx.fill();
}
function rect(ctx, x, y, width, height, localOptions) {
  ctx.beginPath();
  ctx.lineWidth = localOptions.lineWidth;
  if (localOptions.useCurves) {
    const cx2 = (x + x + width) / 2;
    const cy2 = (y + y + height) / 2;
    ctx.ellipse(cx2, cy2, width / 2, height / 2, 0, 0, 2 * Math.PI);
  } else {
    ctx.moveTo(x + localOptions.roundRect, y);
    ctx.lineTo(x + width - localOptions.roundRect, y);
    ctx.quadraticCurveTo(x + width, y, x + width, y + localOptions.roundRect);
    ctx.lineTo(x + width, y + height - localOptions.roundRect);
    ctx.quadraticCurveTo(x + width, y + height, x + width - localOptions.roundRect, y + height);
    ctx.lineTo(x + localOptions.roundRect, y + height);
    ctx.quadraticCurveTo(x, y + height, x, y + height - localOptions.roundRect);
    ctx.lineTo(x, y + localOptions.roundRect);
    ctx.quadraticCurveTo(x, y, x + localOptions.roundRect, y);
    ctx.closePath();
  }
  ctx.stroke();
}
function lines(ctx, points, localOptions) {
  if (points.length < 2)
    return;
  ctx.beginPath();
  ctx.moveTo(points[0][0], points[0][1]);
  for (const pt2 of points) {
    ctx.strokeStyle = colorDepth(pt2[2], localOptions);
    ctx.lineTo(Math.trunc(pt2[0]), Math.trunc(pt2[1]));
  }
  ctx.stroke();
  if (localOptions.fillPolygons) {
    ctx.closePath();
    ctx.fill();
  }
}
function curves(ctx, points, localOptions) {
  if (points.length < 2)
    return;
  ctx.lineWidth = localOptions.lineWidth;
  if (!localOptions.useCurves || points.length <= 2) {
    lines(ctx, points, localOptions);
    return;
  }
  ctx.moveTo(points[0][0], points[0][1]);
  for (let i = 0; i < points.length - 2; i++) {
    const xc = (points[i][0] + points[i + 1][0]) / 2;
    const yc = (points[i][1] + points[i + 1][1]) / 2;
    ctx.quadraticCurveTo(points[i][0], points[i][1], xc, yc);
  }
  ctx.quadraticCurveTo(points[points.length - 2][0], points[points.length - 2][1], points[points.length - 1][0], points[points.length - 1][1]);
  ctx.stroke();
  if (localOptions.fillPolygons) {
    ctx.closePath();
    ctx.fill();
  }
}
function arrow(ctx, from, to2, radius = 5) {
  let angle;
  let x;
  let y;
  ctx.beginPath();
  ctx.moveTo(from[0], from[1]);
  ctx.lineTo(to2[0], to2[1]);
  angle = Math.atan2(to2[1] - from[1], to2[0] - from[0]);
  x = radius * Math.cos(angle) + to2[0];
  y = radius * Math.sin(angle) + to2[1];
  ctx.moveTo(x, y);
  angle += 1 / 3 * (2 * Math.PI);
  x = radius * Math.cos(angle) + to2[0];
  y = radius * Math.sin(angle) + to2[1];
  ctx.lineTo(x, y);
  angle += 1 / 3 * (2 * Math.PI);
  x = radius * Math.cos(angle) + to2[0];
  y = radius * Math.sin(angle) + to2[1];
  ctx.lineTo(x, y);
  ctx.closePath();
  ctx.stroke();
  ctx.fill();
}

// src/draw/options.ts
var options3 = {
  color: "rgba(173, 216, 230, 0.6)",
  labelColor: "rgba(173, 216, 230, 1)",
  shadowColor: "black",
  alpha: 0.5,
  font: 'small-caps 16px "Segoe UI"',
  lineHeight: 18,
  lineWidth: 4,
  pointSize: 2,
  roundRect: 8,
  drawPoints: false,
  drawLabels: true,
  drawBoxes: true,
  drawAttention: true,
  drawGestures: true,
  drawPolygons: true,
  drawGaze: true,
  fillPolygons: false,
  useDepth: true,
  useCurves: false
};

// src/draw/face.ts
var opt;
function drawLabels(f, ctx) {
  if (opt.drawLabels) {
    const labels2 = [];
    labels2.push(`face: ${Math.trunc(100 * f.score)}%`);
    if (f.genderScore)
      labels2.push(`${f.gender || ""} ${Math.trunc(100 * f.genderScore)}%`);
    if (f.age)
      labels2.push(`age: ${f.age || ""}`);
    if (f.iris)
      labels2.push(`distance: ${f.iris}`);
    if (f.real)
      labels2.push(`real: ${Math.trunc(100 * f.real)}%`);
    if (f.live)
      labels2.push(`live: ${Math.trunc(100 * f.live)}%`);
    if (f.emotion && f.emotion.length > 0) {
      const emotion = f.emotion.map((a) => `${Math.trunc(100 * a.score)}% ${a.emotion}`);
      if (emotion.length > 3)
        emotion.length = 3;
      labels2.push(emotion.join(" "));
    }
    if (f.rotation && f.rotation.angle && f.rotation.gaze) {
      if (f.rotation.angle.roll)
        labels2.push(`roll: ${rad2deg(f.rotation.angle.roll)}\xB0 yaw:${rad2deg(f.rotation.angle.yaw)}\xB0 pitch:${rad2deg(f.rotation.angle.pitch)}\xB0`);
      if (f.rotation.gaze.bearing)
        labels2.push(`gaze: ${rad2deg(f.rotation.gaze.bearing)}\xB0`);
    }
    if (labels2.length === 0)
      labels2.push("face");
    ctx.fillStyle = opt.color;
    for (let i = labels2.length - 1; i >= 0; i--) {
      const x = Math.max(f.box[0], 0);
      const y = i * opt.lineHeight + f.box[1];
      if (opt.shadowColor && opt.shadowColor !== "") {
        ctx.fillStyle = opt.shadowColor;
        ctx.fillText(labels2[i], x + 5, y + 16);
      }
      ctx.fillStyle = opt.labelColor;
      ctx.fillText(labels2[i], x + 4, y + 15);
    }
  }
}
function drawIrisElipse(f, ctx) {
  if (f.annotations && f.annotations["leftEyeIris"] && f.annotations["leftEyeIris"][0]) {
    ctx.strokeStyle = opt.useDepth ? "rgba(255, 200, 255, 0.3)" : opt.color;
    ctx.beginPath();
    const sizeX = Math.abs(f.annotations["leftEyeIris"][3][0] - f.annotations["leftEyeIris"][1][0]) / 2;
    const sizeY = Math.abs(f.annotations["leftEyeIris"][4][1] - f.annotations["leftEyeIris"][2][1]) / 2;
    ctx.ellipse(f.annotations["leftEyeIris"][0][0], f.annotations["leftEyeIris"][0][1], sizeX, sizeY, 0, 0, 2 * Math.PI);
    ctx.stroke();
    if (opt.fillPolygons) {
      ctx.fillStyle = opt.useDepth ? "rgba(255, 255, 200, 0.3)" : opt.color;
      ctx.fill();
    }
  }
  if (f.annotations && f.annotations["rightEyeIris"] && f.annotations["rightEyeIris"][0]) {
    ctx.strokeStyle = opt.useDepth ? "rgba(255, 200, 255, 0.3)" : opt.color;
    ctx.beginPath();
    const sizeX = Math.abs(f.annotations["rightEyeIris"][3][0] - f.annotations["rightEyeIris"][1][0]) / 2;
    const sizeY = Math.abs(f.annotations["rightEyeIris"][4][1] - f.annotations["rightEyeIris"][2][1]) / 2;
    ctx.ellipse(f.annotations["rightEyeIris"][0][0], f.annotations["rightEyeIris"][0][1], sizeX, sizeY, 0, 0, 2 * Math.PI);
    ctx.stroke();
    if (opt.fillPolygons) {
      ctx.fillStyle = opt.useDepth ? "rgba(255, 255, 200, 0.3)" : opt.color;
      ctx.fill();
    }
  }
}
function drawGazeSpheres(f, ctx) {
  var _a2;
  if (opt.drawGaze && ((_a2 = f.rotation) == null ? void 0 : _a2.angle) && typeof Path2D !== "undefined") {
    ctx.strokeStyle = "pink";
    const valX = f.box[0] + f.box[2] / 2 - f.box[3] * rad2deg(f.rotation.angle.yaw) / 90;
    const valY = f.box[1] + f.box[3] / 2 + f.box[2] * rad2deg(f.rotation.angle.pitch) / 90;
    const pathV = new Path2D(`
      M ${f.box[0] + f.box[2] / 2} ${f.box[1]}
      C
        ${valX} ${f.box[1]},
        ${valX} ${f.box[1] + f.box[3]},
        ${f.box[0] + f.box[2] / 2} ${f.box[1] + f.box[3]}
    `);
    const pathH = new Path2D(`
      M ${f.box[0]} ${f.box[1] + f.box[3] / 2}
      C 
        ${f.box[0]} ${valY},
        ${f.box[0] + f.box[2]} ${valY},
        ${f.box[0] + f.box[2]} ${f.box[1] + f.box[3] / 2}
    `);
    ctx.stroke(pathH);
    ctx.stroke(pathV);
  }
}
function drawGazeArrows(f, ctx) {
  var _a2, _b2, _c, _d2;
  if (opt.drawGaze && ((_b2 = (_a2 = f.rotation) == null ? void 0 : _a2.gaze) == null ? void 0 : _b2.strength) && ((_d2 = (_c = f.rotation) == null ? void 0 : _c.gaze) == null ? void 0 : _d2.bearing) && f.annotations["leftEyeIris"] && f.annotations["rightEyeIris"] && f.annotations["leftEyeIris"][0] && f.annotations["rightEyeIris"][0]) {
    ctx.strokeStyle = "pink";
    ctx.fillStyle = "pink";
    const leftGaze = [
      f.annotations["leftEyeIris"][0][0] + Math.sin(f.rotation.gaze.bearing) * f.rotation.gaze.strength * f.box[3],
      f.annotations["leftEyeIris"][0][1] + Math.cos(f.rotation.gaze.bearing) * f.rotation.gaze.strength * f.box[2]
    ];
    arrow(ctx, [f.annotations["leftEyeIris"][0][0], f.annotations["leftEyeIris"][0][1]], [leftGaze[0], leftGaze[1]], 4);
    const rightGaze = [
      f.annotations["rightEyeIris"][0][0] + Math.sin(f.rotation.gaze.bearing) * f.rotation.gaze.strength * f.box[3],
      f.annotations["rightEyeIris"][0][1] + Math.cos(f.rotation.gaze.bearing) * f.rotation.gaze.strength * f.box[2]
    ];
    arrow(ctx, [f.annotations["rightEyeIris"][0][0], f.annotations["rightEyeIris"][0][1]], [rightGaze[0], rightGaze[1]], 4);
  }
}
function drawFacePolygons(f, ctx) {
  if (opt.drawPolygons && f.mesh.length >= 468) {
    ctx.lineWidth = 1;
    for (let i = 0; i < TRI468.length / 3; i++) {
      const points = [TRI468[i * 3 + 0], TRI468[i * 3 + 1], TRI468[i * 3 + 2]].map((index2) => f.mesh[index2]);
      lines(ctx, points, opt);
    }
    drawIrisElipse(f, ctx);
  }
}
function drawFacePoints(f, ctx) {
  if (opt.drawPoints && f.mesh.length >= 468) {
    for (let i = 0; i < f.mesh.length; i++) {
      point(ctx, f.mesh[i][0], f.mesh[i][1], f.mesh[i][2], opt);
      if (opt.drawAttention) {
        if (attentionDefinitions.lips.includes(i))
          point(ctx, f.mesh[i][0], f.mesh[i][1], f.mesh[i][2] + 127, opt);
        if (attentionDefinitions.eyeL.includes(i))
          point(ctx, f.mesh[i][0], f.mesh[i][1], f.mesh[i][2] - 127, opt);
        if (attentionDefinitions.eyeR.includes(i))
          point(ctx, f.mesh[i][0], f.mesh[i][1], f.mesh[i][2] - 127, opt);
      }
    }
  }
}
function drawFaceBoxes(f, ctx) {
  if (opt.drawBoxes) {
    rect(ctx, f.box[0], f.box[1], f.box[2], f.box[3], opt);
  }
}
async function face(inCanvas2, result, drawOptions) {
  opt = mergeDeep(options3, drawOptions);
  if (!result || !inCanvas2)
    return;
  const ctx = getCanvasContext(inCanvas2);
  if (!ctx)
    return;
  ctx.font = opt.font;
  ctx.strokeStyle = opt.color;
  ctx.fillStyle = opt.color;
  for (const f of result) {
    drawFaceBoxes(f, ctx);
    drawLabels(f, ctx);
    if (f.mesh && f.mesh.length > 0) {
      drawFacePoints(f, ctx);
      drawFacePolygons(f, ctx);
      drawGazeSpheres(f, ctx);
      drawGazeArrows(f, ctx);
    }
  }
}

// src/draw/body.ts
async function body(inCanvas2, result, drawOptions) {
  var _a2;
  const localOptions = mergeDeep(options3, drawOptions);
  if (!result || !inCanvas2)
    return;
  const ctx = getCanvasContext(inCanvas2);
  if (!ctx)
    return;
  ctx.lineJoin = "round";
  for (let i = 0; i < result.length; i++) {
    ctx.strokeStyle = localOptions.color;
    ctx.fillStyle = localOptions.color;
    ctx.lineWidth = localOptions.lineWidth;
    ctx.font = localOptions.font;
    if (localOptions.drawBoxes && result[i].box && ((_a2 = result[i].box) == null ? void 0 : _a2.length) === 4) {
      rect(ctx, result[i].box[0], result[i].box[1], result[i].box[2], result[i].box[3], localOptions);
      if (localOptions.drawLabels) {
        if (localOptions.shadowColor && localOptions.shadowColor !== "") {
          ctx.fillStyle = localOptions.shadowColor;
          ctx.fillText(`body ${100 * result[i].score}%`, result[i].box[0] + 3, 1 + result[i].box[1] + localOptions.lineHeight, result[i].box[2]);
        }
        ctx.fillStyle = localOptions.labelColor;
        ctx.fillText(`body ${100 * result[i].score}%`, result[i].box[0] + 2, 0 + result[i].box[1] + localOptions.lineHeight, result[i].box[2]);
      }
    }
    if (localOptions.drawPoints && result[i].keypoints) {
      for (let pt2 = 0; pt2 < result[i].keypoints.length; pt2++) {
        if (!result[i].keypoints[pt2].score || result[i].keypoints[pt2].score === 0)
          continue;
        ctx.fillStyle = colorDepth(result[i].keypoints[pt2].position[2], localOptions);
        point(ctx, result[i].keypoints[pt2].position[0], result[i].keypoints[pt2].position[1], 0, localOptions);
      }
    }
    if (localOptions.drawLabels && result[i].keypoints) {
      ctx.font = localOptions.font;
      for (const pt2 of result[i].keypoints) {
        if (!pt2.score || pt2.score === 0)
          continue;
        ctx.fillStyle = colorDepth(pt2.position[2], localOptions);
        ctx.fillText(`${pt2.part} ${Math.trunc(100 * pt2.score)}%`, pt2.position[0] + 4, pt2.position[1] + 4);
      }
    }
    if (localOptions.drawPolygons && result[i].keypoints && result[i].annotations) {
      for (const part of Object.values(result[i].annotations)) {
        for (const connected4 of part)
          curves(ctx, connected4, localOptions);
      }
    }
  }
}

// src/draw/hand.ts
async function hand(inCanvas2, result, drawOptions) {
  const localOptions = mergeDeep(options3, drawOptions);
  if (!result || !inCanvas2)
    return;
  const ctx = getCanvasContext(inCanvas2);
  if (!ctx)
    return;
  ctx.lineJoin = "round";
  ctx.font = localOptions.font;
  for (const h of result) {
    if (localOptions.drawBoxes) {
      ctx.strokeStyle = localOptions.color;
      ctx.fillStyle = localOptions.color;
      rect(ctx, h.box[0], h.box[1], h.box[2], h.box[3], localOptions);
      if (localOptions.drawLabels) {
        if (localOptions.shadowColor && localOptions.shadowColor !== "") {
          ctx.fillStyle = localOptions.shadowColor;
          ctx.fillText(`hand:${Math.trunc(100 * h.score)}%`, h.box[0] + 3, 1 + h.box[1] + localOptions.lineHeight, h.box[2]);
        }
        ctx.fillStyle = localOptions.labelColor;
        ctx.fillText(`hand:${Math.trunc(100 * h.score)}%`, h.box[0] + 2, 0 + h.box[1] + localOptions.lineHeight, h.box[2]);
      }
      ctx.stroke();
    }
    if (localOptions.drawPoints) {
      if (h.keypoints && h.keypoints.length > 0) {
        for (const pt2 of h.keypoints) {
          ctx.fillStyle = colorDepth(pt2[2], localOptions);
          point(ctx, pt2[0], pt2[1], 0, localOptions);
        }
      }
    }
    if (localOptions.drawLabels && h.annotations) {
      const addHandLabel = (part, title) => {
        if (!part || part.length === 0 || !part[0])
          return;
        const z = part[part.length - 1][2] || -256;
        ctx.fillStyle = colorDepth(z, localOptions);
        ctx.fillText(title, part[part.length - 1][0] + 4, part[part.length - 1][1] + 4);
      };
      ctx.font = localOptions.font;
      addHandLabel(h.annotations["index"], "index");
      addHandLabel(h.annotations["middle"], "middle");
      addHandLabel(h.annotations["ring"], "ring");
      addHandLabel(h.annotations["pinky"], "pinky");
      addHandLabel(h.annotations["thumb"], "thumb");
      addHandLabel(h.annotations["palm"], "palm");
    }
    if (localOptions.drawPolygons && h.annotations) {
      const addHandLine = (part) => {
        if (!part || part.length === 0 || !part[0])
          return;
        for (let i = 0; i < part.length; i++) {
          ctx.beginPath();
          const z = part[i][2] || 0;
          ctx.strokeStyle = colorDepth(i * z, localOptions);
          ctx.moveTo(part[i > 0 ? i - 1 : 0][0], part[i > 0 ? i - 1 : 0][1]);
          ctx.lineTo(part[i][0], part[i][1]);
          ctx.stroke();
        }
      };
      ctx.lineWidth = localOptions.lineWidth;
      addHandLine(h.annotations["index"]);
      addHandLine(h.annotations["middle"]);
      addHandLine(h.annotations["ring"]);
      addHandLine(h.annotations["pinky"]);
      addHandLine(h.annotations["thumb"]);
    }
  }
}

// src/draw/object.ts
async function object(inCanvas2, result, drawOptions) {
  const localOptions = mergeDeep(options3, drawOptions);
  if (!result || !inCanvas2)
    return;
  const ctx = getCanvasContext(inCanvas2);
  if (!ctx)
    return;
  ctx.lineJoin = "round";
  ctx.font = localOptions.font;
  for (const h of result) {
    if (localOptions.drawBoxes) {
      ctx.strokeStyle = localOptions.color;
      ctx.fillStyle = localOptions.color;
      rect(ctx, h.box[0], h.box[1], h.box[2], h.box[3], localOptions);
      if (localOptions.drawLabels) {
        const label = `${h.label} ${Math.round(100 * h.score)}%`;
        if (localOptions.shadowColor && localOptions.shadowColor !== "") {
          ctx.fillStyle = localOptions.shadowColor;
          ctx.fillText(label, h.box[0] + 3, 1 + h.box[1] + localOptions.lineHeight, h.box[2]);
        }
        ctx.fillStyle = localOptions.labelColor;
        ctx.fillText(label, h.box[0] + 2, 0 + h.box[1] + localOptions.lineHeight, h.box[2]);
      }
      ctx.stroke();
    }
  }
}

// src/draw/gesture.ts
async function gesture(inCanvas2, result, drawOptions) {
  const localOptions = mergeDeep(options3, drawOptions);
  if (!result || !inCanvas2)
    return;
  if (localOptions.drawGestures) {
    const ctx = getCanvasContext(inCanvas2);
    if (!ctx)
      return;
    ctx.font = localOptions.font;
    ctx.fillStyle = localOptions.color;
    let i = 1;
    for (let j10 = 0; j10 < result.length; j10++) {
      let where = [];
      let what = [];
      [where, what] = Object.entries(result[j10]);
      if (what.length > 1 && what[1].length > 0) {
        const who = where[1] > 0 ? `#${where[1]}` : "";
        const label = `${where[0]} ${who}: ${what[1]}`;
        if (localOptions.shadowColor && localOptions.shadowColor !== "") {
          ctx.fillStyle = localOptions.shadowColor;
          ctx.fillText(label, 8, 2 + i * localOptions.lineHeight);
        }
        ctx.fillStyle = localOptions.labelColor;
        ctx.fillText(label, 6, 0 + i * localOptions.lineHeight);
        i += 1;
      }
    }
  }
}

// src/draw/draw.ts
var drawTime = 0;
async function person(inCanvas2, result, drawOptions) {
  const localOptions = mergeDeep(options3, drawOptions);
  if (!result || !inCanvas2)
    return;
  const ctx = getCanvasContext(inCanvas2);
  if (!ctx)
    return;
  ctx.lineJoin = "round";
  ctx.font = localOptions.font;
  for (let i = 0; i < result.length; i++) {
    if (localOptions.drawBoxes) {
      ctx.strokeStyle = localOptions.color;
      ctx.fillStyle = localOptions.color;
      rect(ctx, result[i].box[0], result[i].box[1], result[i].box[2], result[i].box[3], localOptions);
      if (localOptions.drawLabels) {
        const label = `person #${i}`;
        if (localOptions.shadowColor && localOptions.shadowColor !== "") {
          ctx.fillStyle = localOptions.shadowColor;
          ctx.fillText(label, result[i].box[0] + 3, 1 + result[i].box[1] + localOptions.lineHeight, result[i].box[2]);
        }
        ctx.fillStyle = localOptions.labelColor;
        ctx.fillText(label, result[i].box[0] + 2, 0 + result[i].box[1] + localOptions.lineHeight, result[i].box[2]);
      }
      ctx.stroke();
    }
  }
}
async function canvas2(input, output) {
  if (!input || !output)
    return;
  const ctx = getCanvasContext(output);
  if (!ctx)
    return;
  ctx.drawImage(input, 0, 0);
}
async function all(inCanvas2, result, drawOptions) {
  if (!result || !result.performance || !result || !inCanvas2)
    return null;
  const timeStamp = now();
  const localOptions = mergeDeep(options3, drawOptions);
  const promise = Promise.all([
    face(inCanvas2, result.face, localOptions),
    body(inCanvas2, result.body, localOptions),
    hand(inCanvas2, result.hand, localOptions),
    object(inCanvas2, result.object, localOptions),
    gesture(inCanvas2, result.gesture, localOptions)
  ]);
  drawTime = env.perfadd ? drawTime + Math.round(now() - timeStamp) : Math.round(now() - timeStamp);
  result.performance.draw = drawTime;
  return promise;
}

// src/face/mask.ts
var expandFact = 0.1;
var alpha = 0.5;
function insidePoly(x, y, polygon) {
  let inside = false;
  let j10 = polygon.length - 1;
  for (let i = 0; i < polygon.length; j10 = i++) {
    if (polygon[i].y > y !== polygon[j10].y > y && x < (polygon[j10].x - polygon[i].x) * (y - polygon[i].y) / (polygon[j10].y - polygon[i].y) + polygon[i].x)
      inside = !inside;
  }
  return inside;
}
async function mask(face4) {
  if (!face4.tensor)
    return face4.tensor;
  if (!face4.mesh || face4.mesh.length < 100)
    return face4.tensor;
  const width = face4.tensor.shape[2] || 0;
  const height = face4.tensor.shape[1] || 0;
  const buffer = await face4.tensor.buffer();
  let silhouette = [];
  for (const pt2 of meshAnnotations.silhouette)
    silhouette.push({ x: (face4.mesh[pt2][0] - face4.box[0]) / face4.box[2], y: (face4.mesh[pt2][1] - face4.box[1]) / face4.box[3] });
  if (expandFact && expandFact > 0)
    silhouette = silhouette.map((pt2) => ({ x: pt2.x > 0.5 ? pt2.x + expandFact : pt2.x - expandFact, y: pt2.y > 0.5 ? pt2.y + expandFact : pt2.y - expandFact }));
  for (let x = 0; x < width; x++) {
    for (let y = 0; y < height; y++) {
      const inside = insidePoly(x / width, y / width, silhouette);
      if (!inside) {
        buffer.set(alpha * buffer.get(0, y, x, 0), 0, y, x, 0);
        buffer.set(alpha * buffer.get(0, y, x, 1), 0, y, x, 1);
        buffer.set(alpha * buffer.get(0, y, x, 2), 0, y, x, 2);
      }
    }
  }
  const output = buffer.toTensor();
  De(buffer);
  return output;
}

// src/face/angles.ts
var calculateGaze = (face4) => {
  const radians = (pt1, pt2) => Math.atan2(pt1[1] - pt2[1], pt1[0] - pt2[0]);
  if (!face4.annotations["rightEyeIris"] || !face4.annotations["leftEyeIris"])
    return { bearing: 0, strength: 0 };
  const offsetIris = [0, -0.1];
  const eyeRatio = 1;
  const left = (face4.mesh[33][2] || 0) > (face4.mesh[263][2] || 0);
  const irisCenter = left ? face4.mesh[473] : face4.mesh[468];
  const eyeCenter = left ? [(face4.mesh[133][0] + face4.mesh[33][0]) / 2, (face4.mesh[133][1] + face4.mesh[33][1]) / 2] : [(face4.mesh[263][0] + face4.mesh[362][0]) / 2, (face4.mesh[263][1] + face4.mesh[362][1]) / 2];
  const eyeSize = left ? [face4.mesh[133][0] - face4.mesh[33][0], face4.mesh[23][1] - face4.mesh[27][1]] : [face4.mesh[263][0] - face4.mesh[362][0], face4.mesh[253][1] - face4.mesh[257][1]];
  const eyeDiff = [
    (eyeCenter[0] - irisCenter[0]) / eyeSize[0] - offsetIris[0],
    eyeRatio * (irisCenter[1] - eyeCenter[1]) / eyeSize[1] - offsetIris[1]
  ];
  let strength = Math.sqrt(eyeDiff[0] * eyeDiff[0] + eyeDiff[1] * eyeDiff[1]);
  strength = Math.min(strength, face4.boxRaw[2] / 2, face4.boxRaw[3] / 2);
  const bearing = (radians([0, 0], eyeDiff) + Math.PI / 2) % Math.PI;
  return { bearing, strength };
};
var calculateFaceAngle = (face4, imageSize) => {
  const normalize = (v) => {
    const length = Math.sqrt(v[0] * v[0] + v[1] * v[1] + v[2] * v[2]);
    v[0] /= length;
    v[1] /= length;
    v[2] /= length;
    return v;
  };
  const subVectors = (a, b) => {
    const x = a[0] - b[0];
    const y = a[1] - b[1];
    const z = a[2] - b[2];
    return [x, y, z];
  };
  const crossVectors = (a, b) => {
    const x = a[1] * b[2] - a[2] * b[1];
    const y = a[2] * b[0] - a[0] * b[2];
    const z = a[0] * b[1] - a[1] * b[0];
    return [x, y, z];
  };
  const rotationMatrixToEulerAngle = (r) => {
    const [r00, _r01, _r02, r10, r11, r12, r20, r21, r22] = r;
    let thetaX;
    let thetaY;
    let thetaZ;
    if (r10 < 1) {
      if (r10 > -1) {
        thetaZ = Math.asin(r10);
        thetaY = Math.atan2(-r20, r00);
        thetaX = Math.atan2(-r12, r11);
      } else {
        thetaZ = -Math.PI / 2;
        thetaY = -Math.atan2(r21, r22);
        thetaX = 0;
      }
    } else {
      thetaZ = Math.PI / 2;
      thetaY = Math.atan2(r21, r22);
      thetaX = 0;
    }
    if (isNaN(thetaX))
      thetaX = 0;
    if (isNaN(thetaY))
      thetaY = 0;
    if (isNaN(thetaZ))
      thetaZ = 0;
    return { pitch: 2 * -thetaX, yaw: 2 * -thetaY, roll: 2 * -thetaZ };
  };
  const mesh = face4.meshRaw;
  if (!mesh || mesh.length < 300)
    return { angle: { pitch: 0, yaw: 0, roll: 0 }, matrix: [1, 0, 0, 0, 1, 0, 0, 0, 1], gaze: { bearing: 0, strength: 0 } };
  const size2 = Math.max(face4.boxRaw[2] * imageSize[0], face4.boxRaw[3] * imageSize[1]) / 1.5;
  const pts = [mesh[10], mesh[152], mesh[234], mesh[454]].map((pt2) => [pt2[0] * imageSize[0] / size2, pt2[1] * imageSize[1] / size2, pt2[2]]);
  const y_axis = normalize(subVectors(pts[1], pts[0]));
  let x_axis = normalize(subVectors(pts[3], pts[2]));
  const z_axis = normalize(crossVectors(x_axis, y_axis));
  x_axis = crossVectors(y_axis, z_axis);
  const matrix = [
    x_axis[0],
    x_axis[1],
    x_axis[2],
    y_axis[0],
    y_axis[1],
    y_axis[2],
    z_axis[0],
    z_axis[1],
    z_axis[2]
  ];
  const angle = rotationMatrixToEulerAngle(matrix);
  const gaze = mesh.length === 478 ? calculateGaze(face4) : { bearing: 0, strength: 0 };
  return { angle, matrix, gaze };
};

// src/face/face.ts
var detectFace = async (instance, input) => {
  var _a2, _b2, _c, _d2, _e2, _f, _g2, _h, _i, _j2, _k2, _l2, _m2, _n2, _o2, _p2, _q2, _r2, _s2, _t2, _u2, _v2;
  let timeStamp = now();
  let ageRes;
  let gearRes;
  let genderRes;
  let emotionRes;
  let mobilefacenetRes;
  let antispoofRes;
  let livenessRes;
  let descRes;
  const faceRes = [];
  instance.state = "run:face";
  const faces = await predict10(input, instance.config);
  instance.performance.face = env.perfadd ? (instance.performance.face || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
  if (!input.shape || input.shape.length !== 4)
    return [];
  if (!faces)
    return [];
  for (let i = 0; i < faces.length; i++) {
    instance.analyze("Get Face");
    if (!faces[i].tensor || faces[i].tensor["isDisposedInternal"]) {
      log("Face object is disposed:", faces[i].tensor);
      continue;
    }
    if ((_a2 = instance.config.face.detector) == null ? void 0 : _a2.mask) {
      const masked = await mask(faces[i]);
      De(faces[i].tensor);
      faces[i].tensor = masked;
    }
    const rotation = faces[i].mesh && faces[i].mesh.length > 200 ? calculateFaceAngle(faces[i], [input.shape[2], input.shape[1]]) : null;
    instance.analyze("Start Emotion:");
    if (instance.config.async) {
      emotionRes = ((_b2 = instance.config.face.emotion) == null ? void 0 : _b2.enabled) ? predict8(faces[i].tensor || ms([]), instance.config, i, faces.length) : [];
    } else {
      instance.state = "run:emotion";
      timeStamp = now();
      emotionRes = ((_c = instance.config.face.emotion) == null ? void 0 : _c.enabled) ? await predict8(faces[i].tensor || ms([]), instance.config, i, faces.length) : [];
      instance.performance.emotion = env.perfadd ? (instance.performance.emotion || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
    }
    instance.analyze("End Emotion:");
    instance.analyze("Start AntiSpoof:");
    if (instance.config.async) {
      antispoofRes = ((_d2 = instance.config.face.antispoof) == null ? void 0 : _d2.enabled) ? predict4(faces[i].tensor || ms([]), instance.config, i, faces.length) : 0;
    } else {
      instance.state = "run:antispoof";
      timeStamp = now();
      antispoofRes = ((_e2 = instance.config.face.antispoof) == null ? void 0 : _e2.enabled) ? await predict4(faces[i].tensor || ms([]), instance.config, i, faces.length) : 0;
      instance.performance.antispoof = env.perfadd ? (instance.performance.antispoof || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
    }
    instance.analyze("End AntiSpoof:");
    instance.analyze("Start Liveness:");
    if (instance.config.async) {
      livenessRes = ((_f = instance.config.face.liveness) == null ? void 0 : _f.enabled) ? predict14(faces[i].tensor || ms([]), instance.config, i, faces.length) : 0;
    } else {
      instance.state = "run:liveness";
      timeStamp = now();
      livenessRes = ((_g2 = instance.config.face.liveness) == null ? void 0 : _g2.enabled) ? await predict14(faces[i].tensor || ms([]), instance.config, i, faces.length) : 0;
      instance.performance.liveness = env.perfadd ? (instance.performance.antispoof || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
    }
    instance.analyze("End Liveness:");
    instance.analyze("Start GEAR:");
    if (instance.config.async) {
      gearRes = ((_h = instance.config.face["gear"]) == null ? void 0 : _h.enabled) ? predict(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
    } else {
      instance.state = "run:gear";
      timeStamp = now();
      gearRes = ((_i = instance.config.face["gear"]) == null ? void 0 : _i.enabled) ? await predict(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
      instance.performance.gear = Math.trunc(now() - timeStamp);
    }
    instance.analyze("End GEAR:");
    instance.analyze("Start SSRNet:");
    if (instance.config.async) {
      ageRes = ((_j2 = instance.config.face["ssrnet"]) == null ? void 0 : _j2.enabled) ? predict2(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
      genderRes = ((_k2 = instance.config.face["ssrnet"]) == null ? void 0 : _k2.enabled) ? predict3(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
    } else {
      instance.state = "run:ssrnet";
      timeStamp = now();
      ageRes = ((_l2 = instance.config.face["ssrnet"]) == null ? void 0 : _l2.enabled) ? await predict2(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
      genderRes = ((_m2 = instance.config.face["ssrnet"]) == null ? void 0 : _m2.enabled) ? await predict3(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
      instance.performance.ssrnet = Math.trunc(now() - timeStamp);
    }
    instance.analyze("End SSRNet:");
    instance.analyze("Start MobileFaceNet:");
    if (instance.config.async) {
      mobilefacenetRes = ((_n2 = instance.config.face["mobilefacenet"]) == null ? void 0 : _n2.enabled) ? predict9(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
    } else {
      instance.state = "run:mobilefacenet";
      timeStamp = now();
      mobilefacenetRes = ((_o2 = instance.config.face["mobilefacenet"]) == null ? void 0 : _o2.enabled) ? await predict9(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
      instance.performance.mobilefacenet = Math.trunc(now() - timeStamp);
    }
    instance.analyze("End MobileFaceNet:");
    instance.analyze("Start Description:");
    if (instance.config.async) {
      descRes = ((_p2 = instance.config.face.description) == null ? void 0 : _p2.enabled) ? predict11(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
    } else {
      instance.state = "run:description";
      timeStamp = now();
      descRes = ((_q2 = instance.config.face.description) == null ? void 0 : _q2.enabled) ? await predict11(faces[i].tensor || ms([]), instance.config, i, faces.length) : null;
      instance.performance.description = env.perfadd ? (instance.performance.description || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
    }
    instance.analyze("End Description:");
    if (instance.config.async) {
      [ageRes, genderRes, emotionRes, mobilefacenetRes, descRes, gearRes, antispoofRes, livenessRes] = await Promise.all([ageRes, genderRes, emotionRes, mobilefacenetRes, descRes, gearRes, antispoofRes, livenessRes]);
    }
    instance.analyze("Finish Face:");
    if (((_r2 = instance.config.face["ssrnet"]) == null ? void 0 : _r2.enabled) && ageRes && genderRes) {
      descRes = {
        ...descRes,
        age: ageRes.age,
        gender: genderRes.gender,
        genderScore: genderRes.genderScore
      };
    }
    if (((_s2 = instance.config.face["gear"]) == null ? void 0 : _s2.enabled) && gearRes) {
      descRes = {
        ...descRes,
        age: gearRes.age,
        gender: gearRes.gender,
        genderScore: gearRes.genderScore,
        race: gearRes.race
      };
    }
    if (((_t2 = instance.config.face["mobilefacenet"]) == null ? void 0 : _t2.enabled) && mobilefacenetRes) {
      descRes.descriptor = mobilefacenetRes;
    }
    if (!((_u2 = instance.config.face.iris) == null ? void 0 : _u2.enabled)) {
    }
    const irisSize = faces[i].annotations && faces[i].annotations.leftEyeIris && faces[i].annotations.leftEyeIris[0] && faces[i].annotations.rightEyeIris && faces[i].annotations.rightEyeIris[0] && faces[i].annotations.leftEyeIris.length > 0 && faces[i].annotations.rightEyeIris.length > 0 && faces[i].annotations.leftEyeIris[0] !== null && faces[i].annotations.rightEyeIris[0] !== null ? Math.max(Math.abs(faces[i].annotations.leftEyeIris[3][0] - faces[i].annotations.leftEyeIris[1][0]), Math.abs(faces[i].annotations.rightEyeIris[4][1] - faces[i].annotations.rightEyeIris[2][1])) / input.shape[2] : 0;
    const tensor = ((_v2 = instance.config.face.detector) == null ? void 0 : _v2.return) ? mr(faces[i].tensor) : null;
    De(faces[i].tensor);
    if (faces[i].tensor)
      delete faces[i].tensor;
    const res = {
      ...faces[i],
      id: i
    };
    if (descRes == null ? void 0 : descRes.age)
      res.age = descRes.age;
    if (descRes == null ? void 0 : descRes.gender)
      res.gender = descRes.gender;
    if (descRes == null ? void 0 : descRes.genderScore)
      res.genderScore = descRes == null ? void 0 : descRes.genderScore;
    if (descRes == null ? void 0 : descRes.descriptor)
      res.embedding = descRes == null ? void 0 : descRes.descriptor;
    if (descRes == null ? void 0 : descRes.race)
      res.race = descRes == null ? void 0 : descRes.race;
    if (emotionRes)
      res.emotion = emotionRes;
    if (antispoofRes)
      res.real = antispoofRes;
    if (livenessRes)
      res.live = livenessRes;
    if (irisSize && irisSize !== 0)
      res.iris = Math.trunc(500 / irisSize / 11.7) / 100;
    if (rotation)
      res.rotation = rotation;
    if (tensor)
      res.tensor = tensor;
    faceRes.push(res);
    instance.analyze("End Face");
  }
  instance.analyze("End FaceMesh:");
  if (instance.config.async) {
    if (instance.performance.face)
      delete instance.performance.face;
    if (instance.performance.age)
      delete instance.performance.age;
    if (instance.performance.gender)
      delete instance.performance.gender;
    if (instance.performance.emotion)
      delete instance.performance.emotion;
  }
  return faceRes;
};

// src/gesture/gesture.ts
var body2 = (res) => {
  if (!res)
    return [];
  const gestures = [];
  for (let i = 0; i < res.length; i++) {
    const leftWrist = res[i].keypoints.find((a) => a.part === "leftWrist");
    const rightWrist = res[i].keypoints.find((a) => a.part === "rightWrist");
    const nose = res[i].keypoints.find((a) => a.part === "nose");
    if (nose && leftWrist && rightWrist && leftWrist.position[1] < nose.position[1] && rightWrist.position[1] < nose.position[1])
      gestures.push({ body: i, gesture: "i give up" });
    else if (nose && leftWrist && leftWrist.position[1] < nose.position[1])
      gestures.push({ body: i, gesture: "raise left hand" });
    else if (nose && rightWrist && rightWrist.position[1] < nose.position[1])
      gestures.push({ body: i, gesture: "raise right hand" });
    const leftShoulder = res[i].keypoints.find((a) => a.part === "leftShoulder");
    const rightShoulder = res[i].keypoints.find((a) => a.part === "rightShoulder");
    if (leftShoulder && rightShoulder && Math.abs(leftShoulder.positionRaw[1] - rightShoulder.positionRaw[1]) > 0.1) {
      gestures.push({ body: i, gesture: `leaning ${leftShoulder.position[1] > rightShoulder.position[1] ? "left" : "right"}` });
    }
  }
  return gestures;
};
var face2 = (res) => {
  if (!res)
    return [];
  const gestures = [];
  for (let i = 0; i < res.length; i++) {
    if (res[i].mesh && res[i].mesh.length > 450) {
      const zDiff = (res[i].mesh[33][2] || 0) - (res[i].mesh[263][2] || 0);
      const xDiff = res[i].mesh[33][0] - res[i].mesh[263][0];
      if (Math.abs(zDiff / xDiff) <= 0.15)
        gestures.push({ face: i, gesture: "facing center" });
      else
        gestures.push({ face: i, gesture: `facing ${zDiff < 0 ? "left" : "right"}` });
      const openLeft = Math.abs(res[i].mesh[374][1] - res[i].mesh[386][1]) / Math.abs(res[i].mesh[443][1] - res[i].mesh[450][1]);
      if (openLeft < 0.2)
        gestures.push({ face: i, gesture: "blink left eye" });
      const openRight = Math.abs(res[i].mesh[145][1] - res[i].mesh[159][1]) / Math.abs(res[i].mesh[223][1] - res[i].mesh[230][1]);
      if (openRight < 0.2)
        gestures.push({ face: i, gesture: "blink right eye" });
      const mouthOpen = Math.min(100, 500 * Math.abs(res[i].mesh[13][1] - res[i].mesh[14][1]) / Math.abs(res[i].mesh[10][1] - res[i].mesh[152][1]));
      if (mouthOpen > 10)
        gestures.push({ face: i, gesture: `mouth ${Math.trunc(mouthOpen)}% open` });
      const chinDepth = res[i].mesh[152][2] || 0;
      if (Math.abs(chinDepth) > 10)
        gestures.push({ face: i, gesture: `head ${chinDepth < 0 ? "up" : "down"}` });
    }
  }
  return gestures;
};
var iris = (res) => {
  if (!res)
    return [];
  const gestures = [];
  for (let i = 0; i < res.length; i++) {
    if (!res[i].annotations || !res[i].annotations.leftEyeIris || !res[i].annotations.leftEyeIris[0] || !res[i].annotations.rightEyeIris || !res[i].annotations.rightEyeIris[0])
      continue;
    const sizeXLeft = res[i].annotations.leftEyeIris[3][0] - res[i].annotations.leftEyeIris[1][0];
    const sizeYLeft = res[i].annotations.leftEyeIris[4][1] - res[i].annotations.leftEyeIris[2][1];
    const areaLeft = Math.abs(sizeXLeft * sizeYLeft);
    const sizeXRight = res[i].annotations.rightEyeIris[3][0] - res[i].annotations.rightEyeIris[1][0];
    const sizeYRight = res[i].annotations.rightEyeIris[4][1] - res[i].annotations.rightEyeIris[2][1];
    const areaRight = Math.abs(sizeXRight * sizeYRight);
    let center = false;
    const difference = Math.abs(areaLeft - areaRight) / Math.max(areaLeft, areaRight);
    if (difference < 0.25) {
      center = true;
      gestures.push({ iris: i, gesture: "facing center" });
    }
    const leftIrisCenterX = Math.abs(res[i].mesh[263][0] - res[i].annotations.leftEyeIris[0][0]) / res[i].box[2];
    const rightIrisCenterX = Math.abs(res[i].mesh[33][0] - res[i].annotations.rightEyeIris[0][0]) / res[i].box[2];
    if (leftIrisCenterX > 0.06 || rightIrisCenterX > 0.06)
      center = false;
    if (leftIrisCenterX > rightIrisCenterX) {
      if (leftIrisCenterX > 0.05)
        gestures.push({ iris: i, gesture: "looking right" });
    } else {
      if (rightIrisCenterX > 0.05)
        gestures.push({ iris: i, gesture: "looking left" });
    }
    const rightIrisCenterY = Math.abs(res[i].mesh[145][1] - res[i].annotations.rightEyeIris[0][1]) / res[i].box[3];
    const leftIrisCenterY = Math.abs(res[i].mesh[374][1] - res[i].annotations.leftEyeIris[0][1]) / res[i].box[3];
    if (leftIrisCenterY < 0.01 || rightIrisCenterY < 0.01 || leftIrisCenterY > 0.022 || rightIrisCenterY > 0.022)
      center = false;
    if (leftIrisCenterY < 0.01 || rightIrisCenterY < 0.01)
      gestures.push({ iris: i, gesture: "looking down" });
    if (leftIrisCenterY > 0.022 || rightIrisCenterY > 0.022)
      gestures.push({ iris: i, gesture: "looking up" });
    if (center)
      gestures.push({ iris: i, gesture: "looking center" });
  }
  return gestures;
};
var hand2 = (res) => {
  if (!res)
    return [];
  const gestures = [];
  for (let i = 0; i < res.length; i++) {
    const fingers = [];
    if (res[i]["annotations"]) {
      for (const [finger, pos] of Object.entries(res[i]["annotations"])) {
        if (finger !== "palmBase" && Array.isArray(pos) && pos[0])
          fingers.push({ name: finger.toLowerCase(), position: pos[0] });
      }
    }
    if (fingers && fingers.length > 0) {
      const closest = fingers.reduce((best, a) => (best.position[2] || 0) < (a.position[2] || 0) ? best : a);
      gestures.push({ hand: i, gesture: `${closest.name} forward` });
      const highest = fingers.reduce((best, a) => best.position[1] < a.position[1] ? best : a);
      gestures.push({ hand: i, gesture: `${highest.name} up` });
    }
    if (res[i]["keypoints"]) {
      const poses = match(res[i]["keypoints"]);
      for (const pose of poses)
        gestures.push({ hand: i, gesture: pose.name });
    }
  }
  return gestures;
};

// src/util/interpolate.ts
var bufferedResult = { face: [], body: [], hand: [], gesture: [], object: [], persons: [], performance: {}, timestamp: 0, error: null };
var interpolateTime = 0;
function calc2(newResult, config3) {
  var _a2, _b2, _c, _d2, _e2, _f, _g2, _h, _i, _j2, _k2, _l2, _m2, _n2, _o2, _p2, _q2, _r2, _s2, _t2, _u2, _v2, _w2, _x2, _y2, _z2, _A2;
  const t02 = now();
  if (!newResult)
    return { face: [], body: [], hand: [], gesture: [], object: [], persons: [], performance: {}, timestamp: 0, error: null };
  const elapsed = Date.now() - newResult.timestamp;
  const bufferedFactor = elapsed < 1e3 ? 8 - Math.log(elapsed + 1) : 1;
  if (newResult.canvas)
    bufferedResult.canvas = newResult.canvas;
  if (newResult.error)
    bufferedResult.error = newResult.error;
  if (!bufferedResult.body || newResult.body.length !== bufferedResult.body.length) {
    bufferedResult.body = JSON.parse(JSON.stringify(newResult.body));
  } else {
    for (let i = 0; i < newResult.body.length; i++) {
      const box = newResult.body[i].box.map((newBoxCoord, j10) => ((bufferedFactor - 1) * bufferedResult.body[i].box[j10] + newBoxCoord) / bufferedFactor);
      const boxRaw = newResult.body[i].boxRaw.map((newBoxCoord, j10) => ((bufferedFactor - 1) * bufferedResult.body[i].boxRaw[j10] + newBoxCoord) / bufferedFactor);
      const keypoints = newResult.body[i].keypoints.map((newKpt, j10) => {
        var _a3, _b3, _c2, _d3, _e3, _f2, _g3, _h2, _i2;
        return {
          score: newKpt.score,
          part: newKpt.part,
          position: [
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (bufferedResult.body[i].keypoints[j10].position[0] || 0) + (newKpt.position[0] || 0)) / bufferedFactor : newKpt.position[0],
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (bufferedResult.body[i].keypoints[j10].position[1] || 0) + (newKpt.position[1] || 0)) / bufferedFactor : newKpt.position[1],
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (bufferedResult.body[i].keypoints[j10].position[2] || 0) + (newKpt.position[2] || 0)) / bufferedFactor : newKpt.position[2]
          ],
          positionRaw: [
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (bufferedResult.body[i].keypoints[j10].positionRaw[0] || 0) + (newKpt.positionRaw[0] || 0)) / bufferedFactor : newKpt.positionRaw[0],
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (bufferedResult.body[i].keypoints[j10].positionRaw[1] || 0) + (newKpt.positionRaw[1] || 0)) / bufferedFactor : newKpt.positionRaw[1],
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (bufferedResult.body[i].keypoints[j10].positionRaw[2] || 0) + (newKpt.positionRaw[2] || 0)) / bufferedFactor : newKpt.positionRaw[2]
          ],
          distance: [
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (((_a3 = bufferedResult.body[i].keypoints[j10].distance) == null ? void 0 : _a3[0]) || 0) + (((_b3 = newKpt.distance) == null ? void 0 : _b3[0]) || 0)) / bufferedFactor : (_c2 = newKpt.distance) == null ? void 0 : _c2[0],
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (((_d3 = bufferedResult.body[i].keypoints[j10].distance) == null ? void 0 : _d3[1]) || 0) + (((_e3 = newKpt.distance) == null ? void 0 : _e3[1]) || 0)) / bufferedFactor : (_f2 = newKpt.distance) == null ? void 0 : _f2[1],
            bufferedResult.body[i].keypoints[j10] ? ((bufferedFactor - 1) * (((_g3 = bufferedResult.body[i].keypoints[j10].distance) == null ? void 0 : _g3[2]) || 0) + (((_h2 = newKpt.distance) == null ? void 0 : _h2[2]) || 0)) / bufferedFactor : (_i2 = newKpt.distance) == null ? void 0 : _i2[2]
          ]
        };
      });
      const annotations2 = {};
      let coords = { connected: {} };
      if ((_b2 = (_a2 = config3.body) == null ? void 0 : _a2.modelPath) == null ? void 0 : _b2.includes("efficientpose"))
        coords = efficientposecoords_exports;
      else if ((_d2 = (_c = config3.body) == null ? void 0 : _c.modelPath) == null ? void 0 : _d2.includes("blazepose"))
        coords = blazeposecoords_exports;
      else if ((_f = (_e2 = config3.body) == null ? void 0 : _e2.modelPath) == null ? void 0 : _f.includes("movenet"))
        coords = movenetcoords_exports;
      for (const [name, indexes] of Object.entries(coords.connected)) {
        const pt2 = [];
        for (let j10 = 0; j10 < indexes.length - 1; j10++) {
          const pt0 = keypoints.find((kp2) => kp2.part === indexes[j10]);
          const pt1 = keypoints.find((kp2) => kp2.part === indexes[j10 + 1]);
          if (pt0 && pt1)
            pt2.push([pt0.position, pt1.position]);
        }
        annotations2[name] = pt2;
      }
      bufferedResult.body[i] = { ...newResult.body[i], box, boxRaw, keypoints, annotations: annotations2 };
    }
  }
  if (!bufferedResult.hand || newResult.hand.length !== bufferedResult.hand.length) {
    bufferedResult.hand = JSON.parse(JSON.stringify(newResult.hand));
  } else {
    for (let i = 0; i < newResult.hand.length; i++) {
      const box = newResult.hand[i].box.map((b, j10) => ((bufferedFactor - 1) * bufferedResult.hand[i].box[j10] + b) / bufferedFactor);
      const boxRaw = newResult.hand[i].boxRaw.map((b, j10) => ((bufferedFactor - 1) * bufferedResult.hand[i].boxRaw[j10] + b) / bufferedFactor);
      if (bufferedResult.hand[i].keypoints.length !== newResult.hand[i].keypoints.length)
        bufferedResult.hand[i].keypoints = newResult.hand[i].keypoints;
      const keypoints = newResult.hand[i].keypoints && newResult.hand[i].keypoints.length > 0 ? newResult.hand[i].keypoints.map((landmark, j10) => landmark.map((coord, k) => ((bufferedFactor - 1) * (bufferedResult.hand[i].keypoints[j10][k] || 1) + (coord || 0)) / bufferedFactor)) : [];
      let annotations2 = {};
      if (Object.keys(bufferedResult.hand[i].annotations).length !== Object.keys(newResult.hand[i].annotations).length) {
        bufferedResult.hand[i].annotations = newResult.hand[i].annotations;
        annotations2 = bufferedResult.hand[i].annotations;
      } else if (newResult.hand[i].annotations) {
        for (const key of Object.keys(newResult.hand[i].annotations)) {
          annotations2[key] = newResult.hand[i].annotations[key] && newResult.hand[i].annotations[key][0] ? newResult.hand[i].annotations[key].map((val, j10) => val.map((coord, k) => ((bufferedFactor - 1) * bufferedResult.hand[i].annotations[key][j10][k] + coord) / bufferedFactor)) : null;
        }
      }
      bufferedResult.hand[i] = { ...newResult.hand[i], box, boxRaw, keypoints, annotations: annotations2 };
    }
  }
  if (!bufferedResult.face || newResult.face.length !== bufferedResult.face.length) {
    bufferedResult.face = JSON.parse(JSON.stringify(newResult.face));
  } else {
    for (let i = 0; i < newResult.face.length; i++) {
      const box = newResult.face[i].box.map((b, j10) => ((bufferedFactor - 1) * bufferedResult.face[i].box[j10] + b) / bufferedFactor);
      const boxRaw = newResult.face[i].boxRaw.map((b, j10) => ((bufferedFactor - 1) * bufferedResult.face[i].boxRaw[j10] + b) / bufferedFactor);
      if (newResult.face[i].rotation) {
        const rotation = { matrix: [0, 0, 0, 0, 0, 0, 0, 0, 0], angle: { roll: 0, yaw: 0, pitch: 0 }, gaze: { bearing: 0, strength: 0 } };
        rotation.matrix = (_g2 = newResult.face[i].rotation) == null ? void 0 : _g2.matrix;
        rotation.angle = {
          roll: ((bufferedFactor - 1) * (((_i = (_h = bufferedResult.face[i].rotation) == null ? void 0 : _h.angle) == null ? void 0 : _i.roll) || 0) + (((_k2 = (_j2 = newResult.face[i].rotation) == null ? void 0 : _j2.angle) == null ? void 0 : _k2.roll) || 0)) / bufferedFactor,
          yaw: ((bufferedFactor - 1) * (((_m2 = (_l2 = bufferedResult.face[i].rotation) == null ? void 0 : _l2.angle) == null ? void 0 : _m2.yaw) || 0) + (((_o2 = (_n2 = newResult.face[i].rotation) == null ? void 0 : _n2.angle) == null ? void 0 : _o2.yaw) || 0)) / bufferedFactor,
          pitch: ((bufferedFactor - 1) * (((_q2 = (_p2 = bufferedResult.face[i].rotation) == null ? void 0 : _p2.angle) == null ? void 0 : _q2.pitch) || 0) + (((_s2 = (_r2 = newResult.face[i].rotation) == null ? void 0 : _r2.angle) == null ? void 0 : _s2.pitch) || 0)) / bufferedFactor
        };
        rotation.gaze = {
          bearing: ((bufferedFactor - 1) * (((_u2 = (_t2 = bufferedResult.face[i].rotation) == null ? void 0 : _t2.gaze) == null ? void 0 : _u2.bearing) || 0) + (((_w2 = (_v2 = newResult.face[i].rotation) == null ? void 0 : _v2.gaze) == null ? void 0 : _w2.bearing) || 0)) / bufferedFactor,
          strength: ((bufferedFactor - 1) * (((_y2 = (_x2 = bufferedResult.face[i].rotation) == null ? void 0 : _x2.gaze) == null ? void 0 : _y2.strength) || 0) + (((_A2 = (_z2 = newResult.face[i].rotation) == null ? void 0 : _z2.gaze) == null ? void 0 : _A2.strength) || 0)) / bufferedFactor
        };
        bufferedResult.face[i] = { ...newResult.face[i], rotation, box, boxRaw };
      }
      bufferedResult.face[i] = { ...newResult.face[i], box, boxRaw };
    }
  }
  if (!bufferedResult.object || newResult.object.length !== bufferedResult.object.length) {
    bufferedResult.object = JSON.parse(JSON.stringify(newResult.object));
  } else {
    for (let i = 0; i < newResult.object.length; i++) {
      const box = newResult.object[i].box.map((b, j10) => ((bufferedFactor - 1) * bufferedResult.object[i].box[j10] + b) / bufferedFactor);
      const boxRaw = newResult.object[i].boxRaw.map((b, j10) => ((bufferedFactor - 1) * bufferedResult.object[i].boxRaw[j10] + b) / bufferedFactor);
      bufferedResult.object[i] = { ...newResult.object[i], box, boxRaw };
    }
  }
  if (newResult.persons) {
    const newPersons = newResult.persons;
    if (!bufferedResult.persons || newPersons.length !== bufferedResult.persons.length) {
      bufferedResult.persons = JSON.parse(JSON.stringify(newPersons));
    } else {
      for (let i = 0; i < newPersons.length; i++) {
        bufferedResult.persons[i].box = newPersons[i].box.map((box, j10) => ((bufferedFactor - 1) * bufferedResult.persons[i].box[j10] + box) / bufferedFactor);
      }
    }
  }
  if (newResult.gesture)
    bufferedResult.gesture = newResult.gesture;
  const t12 = now();
  interpolateTime = env.perfadd ? interpolateTime + Math.round(t12 - t02) : Math.round(t12 - t02);
  if (newResult.performance)
    bufferedResult.performance = { ...newResult.performance, interpolate: interpolateTime };
  return bufferedResult;
}

// src/face/match.ts
var match_exports = {};
__export(match_exports, {
  distance: () => distance,
  match: () => match2,
  similarity: () => similarity
});
function distance(descriptor1, descriptor2, options4 = { order: 2, multiplier: 25 }) {
  let sum = 0;
  for (let i = 0; i < descriptor1.length; i++) {
    const diff = !options4.order || options4.order === 2 ? descriptor1[i] - descriptor2[i] : Math.abs(descriptor1[i] - descriptor2[i]);
    sum += !options4.order || options4.order === 2 ? diff * diff : diff ** options4.order;
  }
  return (options4.multiplier || 20) * sum;
}
var normalizeDistance = (dist, order, min, max) => {
  if (dist === 0)
    return 1;
  const root = order === 2 ? Math.sqrt(dist) : dist ** (1 / order);
  const norm = (1 - root / 100 - min) / (max - min);
  const clamp2 = Math.max(Math.min(norm, 1), 0);
  return clamp2;
};
function similarity(descriptor1, descriptor2, options4 = { order: 2, multiplier: 25, min: 0.2, max: 0.8 }) {
  const dist = distance(descriptor1, descriptor2, options4);
  return normalizeDistance(dist, options4.order || 2, options4.min || 0, options4.max || 1);
}
function match2(descriptor, descriptors, options4 = { order: 2, multiplier: 25, threshold: 0, min: 0.2, max: 0.8 }) {
  if (!Array.isArray(descriptor) || !Array.isArray(descriptors) || descriptor.length < 64 || descriptors.length === 0 || descriptor.length !== descriptors[0].length) {
    return { index: -1, distance: Number.POSITIVE_INFINITY, similarity: 0 };
  }
  let lowestDistance = Number.MAX_SAFE_INTEGER;
  let index2 = -1;
  for (let i = 0; i < descriptors.length; i++) {
    const res = distance(descriptor, descriptors[i], options4);
    if (res < lowestDistance) {
      lowestDistance = res;
      index2 = i;
    }
    if (lowestDistance < (options4.threshold || 0))
      break;
  }
  const normalizedSimilarity = normalizeDistance(lowestDistance, options4.order || 2, options4.min || 0, options4.max || 1);
  return { index: index2, distance: lowestDistance, similarity: normalizedSimilarity };
}

// src/util/persons.ts
function join2(faces, bodies, hands, gestures, shape) {
  var _a2, _b2, _c, _d2, _e2, _f, _g2, _h, _i, _j2, _k2, _l2, _m2, _n2, _o2, _p2;
  let id2 = 0;
  const persons = [];
  for (const face4 of faces) {
    const person2 = { id: id2++, face: face4, body: null, hands: { left: null, right: null }, gestures: [], box: [0, 0, 0, 0] };
    for (const body4 of bodies) {
      if (face4.box[0] > body4.box[0] && face4.box[0] < body4.box[0] + body4.box[2] && face4.box[1] + face4.box[3] > body4.box[1] && face4.box[1] + face4.box[3] < body4.box[1] + body4.box[3]) {
        person2.body = body4;
      }
    }
    if (person2.body) {
      for (const hand3 of hands) {
        if (hand3.box[0] + hand3.box[2] > person2.body.box[0] && hand3.box[0] + hand3.box[2] < person2.body.box[0] + person2.body.box[2] && hand3.box[1] + hand3.box[3] > person2.body.box[1] && hand3.box[1] + hand3.box[3] < person2.body.box[1] + person2.body.box[3]) {
          if (person2.hands)
            person2.hands.left = hand3;
        }
        if (hand3.box[0] < person2.body.box[0] + person2.body.box[2] && hand3.box[0] > person2.body.box[0] && hand3.box[1] + hand3.box[3] > person2.body.box[1] && hand3.box[1] + hand3.box[3] < person2.body.box[1] + person2.body.box[3]) {
          if (person2.hands)
            person2.hands.right = hand3;
        }
      }
    }
    for (const gesture2 of gestures) {
      if (gesture2["face"] !== void 0 && gesture2["face"] === face4.id)
        (_a2 = person2.gestures) == null ? void 0 : _a2.push(gesture2);
      else if (gesture2["iris"] !== void 0 && gesture2["iris"] === face4.id)
        (_b2 = person2.gestures) == null ? void 0 : _b2.push(gesture2);
      else if (gesture2["body"] !== void 0 && gesture2["body"] === ((_c = person2.body) == null ? void 0 : _c.id))
        (_d2 = person2.gestures) == null ? void 0 : _d2.push(gesture2);
      else if (gesture2["hand"] !== void 0 && gesture2["hand"] === ((_f = (_e2 = person2.hands) == null ? void 0 : _e2.left) == null ? void 0 : _f.id))
        (_g2 = person2.gestures) == null ? void 0 : _g2.push(gesture2);
      else if (gesture2["hand"] !== void 0 && gesture2["hand"] === ((_i = (_h = person2.hands) == null ? void 0 : _h.right) == null ? void 0 : _i.id))
        (_j2 = person2.gestures) == null ? void 0 : _j2.push(gesture2);
    }
    const x = [];
    const y = [];
    const extractXY = (box) => {
      if (box && box.length === 4) {
        x.push(box[0], box[0] + box[2]);
        y.push(box[1], box[1] + box[3]);
      }
    };
    extractXY((_k2 = person2.face) == null ? void 0 : _k2.box);
    extractXY((_l2 = person2.body) == null ? void 0 : _l2.box);
    extractXY((_n2 = (_m2 = person2.hands) == null ? void 0 : _m2.left) == null ? void 0 : _n2.box);
    extractXY((_p2 = (_o2 = person2.hands) == null ? void 0 : _o2.right) == null ? void 0 : _p2.box);
    const minX = Math.min(...x);
    const minY = Math.min(...y);
    person2.box = [minX, minY, Math.max(...x) - minX, Math.max(...y) - minY];
    if (shape && shape[1] && shape[2])
      person2.boxRaw = [person2.box[0] / shape[2], person2.box[1] / shape[1], person2.box[2] / shape[2], person2.box[3] / shape[1]];
    persons.push(person2);
  }
  return persons;
}

// src/sample.ts
var face3 = `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==`;
var body3 = `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`;

// src/warmup.ts
async function warmupBitmap(instance) {
  const b64toBlob = (base64, type = "application/octet-stream") => fetch(`data:${type};base64,${base64}`).then((res2) => res2.blob());
  let blob;
  let res;
  switch (instance.config.warmup) {
    case "face":
      blob = await b64toBlob(face3);
      break;
    case "body":
    case "full":
      blob = await b64toBlob(body3);
      break;
    default:
      blob = null;
  }
  if (blob) {
    const bitmap = await createImageBitmap(blob);
    res = await instance.detect(bitmap, instance.config);
    bitmap.close();
  }
  return res;
}
async function warmupCanvas(instance) {
  return new Promise((resolve) => {
    let src;
    switch (instance.config.warmup) {
      case "face":
        src = "data:image/jpeg;base64," + face3;
        break;
      case "full":
      case "body":
        src = "data:image/jpeg;base64," + body3;
        break;
      default:
        src = null;
    }
    let img;
    if (typeof Image !== "undefined")
      img = new Image();
    else if (env.Image)
      img = new env.Image();
    else
      return;
    img.onload = async () => {
      const canvas3 = canvas(img.naturalWidth, img.naturalHeight);
      if (!canvas3) {
        log("Warmup: Canvas not found");
        resolve(void 0);
      } else {
        const ctx = canvas3.getContext("2d");
        if (ctx)
          ctx.drawImage(img, 0, 0);
        const tensor = await instance.image(canvas3);
        const res = await instance.detect(tensor.tensor, instance.config);
        resolve(res);
      }
    };
    if (src)
      img.src = src;
    else
      resolve(void 0);
  });
}
async function warmupNode(instance) {
  const atob2 = (str) => Buffer.from(str, "base64");
  let img;
  if (instance.config.warmup === "face")
    img = atob2(face3);
  else
    img = atob2(body3);
  let res;
  if ("node" in tfjs_esm_exports) {
    const data = (void 0).decodeJpeg(img);
    const expanded = data.expandDims(0);
    instance.tf.dispose(data);
    res = await instance.detect(expanded, instance.config);
    instance.tf.dispose(expanded);
  } else {
    if (instance.config.debug)
      log("Warmup tfjs-node not loaded");
  }
  return res;
}
async function runInference(instance) {
  let res;
  if (typeof createImageBitmap === "function")
    res = await warmupBitmap(instance);
  else if (typeof Image !== "undefined" || env.Canvas !== void 0)
    res = await warmupCanvas(instance);
  else
    res = await warmupNode(instance);
  return res;
}
async function runCompile(allModels) {
  const backendType = vpe();
  const webGLBackend = HA();
  if (backendType !== "webgl" && backendType !== "humangl" || (!webGLBackend || !webGLBackend.checkCompileCompletion)) {
    return;
  }
  K().set("ENGINE_COMPILE_ONLY", true);
  const numTensorsStart = ds().state.numTensors;
  const compiledModels = [];
  for (const [modelName, model18] of Object.entries(allModels).filter(([key, val]) => key !== null && val !== null)) {
    const shape = model18.inputs && model18.inputs[0] && model18.inputs[0].shape ? [...model18.inputs[0].shape] : [1, 64, 64, 3];
    const dtype = model18.inputs && model18.inputs[0] && model18.inputs[0].dtype ? model18.inputs[0].dtype : "float32";
    for (let dim = 0; dim < shape.length; dim++) {
      if (shape[dim] === -1)
        shape[dim] = dim === 0 ? 1 : 64;
    }
    const tensor = $t(shape, dtype);
    try {
      const res = model18.execute(tensor);
      compiledModels.push(modelName);
      if (Array.isArray(res))
        res.forEach((t) => De(t));
      else
        De(res);
    } catch (e) {
      log("compile fail model:", modelName);
    }
    De(tensor);
  }
  const kernels = await webGLBackend.checkCompileCompletionAsync();
  webGLBackend.getUniformLocations();
  log("compile pass models:", compiledModels);
  log("compile pass kernels:", kernels.length);
  K().set("ENGINE_COMPILE_ONLY", false);
  const numTensorsEnd = ds().state.numTensors;
  if (numTensorsEnd - numTensorsStart > 0)
    log("tensor leak:", numTensorsEnd - numTensorsStart);
}
async function warmup(instance, userConfig) {
  const t02 = now();
  instance.state = "warmup";
  if (userConfig)
    instance.config = mergeDeep(instance.config, userConfig);
  if (!instance.config.warmup || instance.config.warmup.length === 0 || instance.config.warmup === "none") {
    return { face: [], body: [], hand: [], gesture: [], object: [], performance: instance.performance, timestamp: now(), persons: [], error: null };
  }
  return new Promise(async (resolve) => {
    await runCompile(instance.models);
    const res = await runInference(instance);
    const t12 = now();
    if (instance.config.debug)
      log("warmup", instance.config.warmup, Math.round(t12 - t02), "ms");
    instance.emit("warmup");
    resolve(res);
  });
}

// src/human.ts
var _numTensors, _analyzeMemoryLeaks, _checkSanity, _sanity;
var Human = class {
  constructor(userConfig) {
    __publicField(this, "version");
    __publicField(this, "config");
    __publicField(this, "result");
    __publicField(this, "state");
    __publicField(this, "process");
    __publicField(this, "tf");
    __publicField(this, "env");
    __publicField(this, "draw");
    __publicField(this, "models");
    __publicField(this, "events");
    __publicField(this, "faceTriangulation");
    __publicField(this, "faceUVMap");
    __publicField(this, "performance");
    __privateAdd(this, _numTensors, void 0);
    __privateAdd(this, _analyzeMemoryLeaks, void 0);
    __privateAdd(this, _checkSanity, void 0);
    __publicField(this, "gl");
    __publicField(this, "analyze", (...msg) => {
      if (!__privateGet(this, _analyzeMemoryLeaks))
        return;
      const currentTensors = this.tf.engine().state.numTensors;
      const previousTensors = __privateGet(this, _numTensors);
      __privateSet(this, _numTensors, currentTensors);
      const leaked = currentTensors - previousTensors;
      if (leaked !== 0)
        log(...msg, leaked);
    });
    __privateAdd(this, _sanity, (input) => {
      if (!__privateGet(this, _checkSanity))
        return null;
      if (!input)
        return "input is not defined";
      if (this.env.node && !(input instanceof et))
        return "input must be a tensor";
      try {
        this.tf.getBackend();
      } catch (e) {
        return "backend not loaded";
      }
      return null;
    });
    __publicField(this, "similarity", similarity);
    __publicField(this, "distance", distance);
    __publicField(this, "match", match2);
    __publicField(this, "emit", (event) => {
      var _a2;
      if (this.events && this.events.dispatchEvent)
        (_a2 = this.events) == null ? void 0 : _a2.dispatchEvent(new Event(event));
    });
    this.env = env;
    config.wasmPath = khe["tfjs-core"].includes("-") ? "https://vladmandic.github.io/tfjs/dist/" : `https://cdn.jsdelivr.net/npm/@tensorflow/tfjs-backend-wasm@${cpe}/dist/`;
    config.modelBasePath = env.browser ? "../models/" : "file://models/";
    config.backend = env.browser ? "humangl" : "tensorflow";
    this.version = version;
    Object.defineProperty(this, "version", { value: version });
    this.config = JSON.parse(JSON.stringify(config));
    Object.seal(this.config);
    this.config.cacheModels = typeof indexedDB !== "undefined";
    if (userConfig)
      this.config = mergeDeep(this.config, userConfig);
    setModelLoadOptions(this.config);
    this.tf = tfjs_esm_exports;
    this.state = "idle";
    __privateSet(this, _numTensors, 0);
    __privateSet(this, _analyzeMemoryLeaks, false);
    __privateSet(this, _checkSanity, false);
    this.performance = {};
    this.events = typeof EventTarget !== "undefined" ? new EventTarget() : void 0;
    this.models = new Models();
    this.draw = {
      options: options3,
      canvas: (input, output) => canvas2(input, output),
      face: (output, result, options4) => face(output, result, options4),
      body: (output, result, options4) => body(output, result, options4),
      hand: (output, result, options4) => hand(output, result, options4),
      gesture: (output, result, options4) => gesture(output, result, options4),
      object: (output, result, options4) => object(output, result, options4),
      person: (output, result, options4) => person(output, result, options4),
      all: (output, result, options4) => all(output, result, options4)
    };
    this.result = { face: [], body: [], hand: [], gesture: [], object: [], performance: {}, timestamp: 0, persons: [], error: null };
    this.process = { tensor: null, canvas: null };
    this.faceTriangulation = triangulation;
    this.faceUVMap = uvmap;
    this.gl = config2;
    this.emit("create");
  }
  reset() {
    const currentBackend = this.config.backend;
    this.config = JSON.parse(JSON.stringify(config));
    this.config.backend = currentBackend;
  }
  validate(userConfig) {
    return validate(config, userConfig || this.config);
  }
  now() {
    return now();
  }
  image(input, getTensor = true) {
    return process2(input, this.config, getTensor);
  }
  async segmentation(input, background) {
    return process5(input, background, this.config);
  }
  enhance(input) {
    return enhance(input);
  }
  compare(firstImageTensor, secondImageTensor) {
    return compare(this.config, firstImageTensor, secondImageTensor);
  }
  async init() {
    await check(this, true);
    await this.tf.ready();
  }
  async load(userConfig) {
    this.state = "load";
    const timeStamp = now();
    const count2 = Object.values(this.models).filter((model18) => model18).length;
    if (userConfig)
      this.config = mergeDeep(this.config, userConfig);
    if (this.env.initial) {
      if (this.config.debug)
        log(`version: ${this.version}`);
      if (this.config.debug)
        log(`tfjs version: ${this.tf.version["tfjs-core"]}`);
      if (!await check(this))
        log("error: backend check failed");
      await ype();
      if (this.env.browser) {
        if (this.config.debug)
          log("configuration:", this.config);
        if (this.config.debug)
          log("environment:", this.env);
        if (this.config.debug)
          log("tf flags:", this.tf.ENV["flags"]);
      }
    }
    await load19(this);
    if (this.env.initial && this.config.debug)
      log("tf engine state:", this.tf.engine().state.numBytes, "bytes", this.tf.engine().state.numTensors, "tensors");
    this.env.initial = false;
    const loaded = Object.values(this.models).filter((model18) => model18).length;
    if (loaded !== count2) {
      await validate2(this);
      this.emit("load");
    }
    const current = Math.trunc(now() - timeStamp);
    if (current > (this.performance.loadModels || 0))
      this.performance.loadModels = this.env.perfadd ? (this.performance.loadModels || 0) + current : current;
  }
  next(result = this.result) {
    return calc2(result, this.config);
  }
  async warmup(userConfig) {
    const t02 = now();
    const res = await warmup(this, userConfig);
    const t12 = now();
    this.performance.warmup = Math.trunc(t12 - t02);
    return res;
  }
  async profile(input, userConfig) {
    const profile = await this.tf.profile(() => this.detect(input, userConfig));
    const kernels = {};
    for (const kernel of profile.kernels) {
      if (kernels[kernel.name])
        kernels[kernel.name] += kernel.kernelTimeMs;
      else
        kernels[kernel.name] = kernel.kernelTimeMs;
    }
    const kernelArr = [];
    Object.entries(kernels).forEach((key) => kernelArr.push({ name: key[0], ms: key[1] }));
    kernelArr.sort((a, b) => b.ms - a.ms);
    kernelArr.length = 20;
    const res = {};
    for (const kernel of kernelArr)
      res[kernel.name] = kernel.ms;
    return res;
  }
  async detect(input, userConfig) {
    this.state = "detect";
    return new Promise(async (resolve) => {
      var _a2, _b2, _c, _d2, _e2, _f, _g2, _h, _i, _j2, _k2, _l2, _m2, _n2, _o2, _p2, _q2, _r2, _s2, _t2, _u2, _v2;
      this.state = "config";
      let timeStamp;
      this.config = mergeDeep(this.config, userConfig);
      this.state = "check";
      const error = __privateGet(this, _sanity).call(this, input);
      if (error) {
        log(error, input);
        this.emit("error");
        resolve({ face: [], body: [], hand: [], gesture: [], object: [], performance: this.performance, timestamp: now(), persons: [], error });
      }
      const timeStart = now();
      await check(this);
      await this.load();
      timeStamp = now();
      this.state = "image";
      const img = await process2(input, this.config);
      this.process = img;
      this.performance.inputProcess = this.env.perfadd ? (this.performance.inputProcess || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
      this.analyze("Get Image:");
      if (!img.tensor) {
        if (this.config.debug)
          log("could not convert input to tensor");
        this.emit("error");
        resolve({ face: [], body: [], hand: [], gesture: [], object: [], performance: this.performance, timestamp: now(), persons: [], error: "could not convert input to tensor" });
        return;
      }
      this.emit("image");
      timeStamp = now();
      this.config.skipAllowed = await skip(this.config, img.tensor);
      if (!this.performance.totalFrames)
        this.performance.totalFrames = 0;
      if (!this.performance.cachedFrames)
        this.performance.cachedFrames = 0;
      this.performance.totalFrames++;
      if (this.config.skipAllowed)
        this.performance.cachedFrames++;
      this.performance.cacheCheck = this.env.perfadd ? (this.performance.cacheCheck || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
      this.analyze("Check Changed:");
      let faceRes = [];
      let bodyRes = [];
      let handRes = [];
      let objectRes = [];
      this.state = "detect:face";
      if (this.config.async) {
        faceRes = this.config.face.enabled ? detectFace(this, img.tensor) : [];
        if (this.performance.face)
          delete this.performance.face;
      } else {
        timeStamp = now();
        faceRes = this.config.face.enabled ? await detectFace(this, img.tensor) : [];
        this.performance.face = this.env.perfadd ? (this.performance.face || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
      }
      if (this.config.async && (this.config.body.maxDetected === -1 || this.config.hand.maxDetected === -1))
        faceRes = await faceRes;
      this.analyze("Start Body:");
      this.state = "detect:body";
      const bodyConfig = this.config.body.maxDetected === -1 ? mergeDeep(this.config, { body: { maxDetected: this.config.face.enabled ? 1 * faceRes.length : 1 } }) : this.config;
      if (this.config.async) {
        if ((_a2 = this.config.body.modelPath) == null ? void 0 : _a2.includes("posenet"))
          bodyRes = this.config.body.enabled ? predict17(img.tensor, bodyConfig) : [];
        else if ((_b2 = this.config.body.modelPath) == null ? void 0 : _b2.includes("blazepose"))
          bodyRes = this.config.body.enabled ? predict5(img.tensor, bodyConfig) : [];
        else if ((_c = this.config.body.modelPath) == null ? void 0 : _c.includes("efficientpose"))
          bodyRes = this.config.body.enabled ? predict7(img.tensor, bodyConfig) : [];
        else if ((_d2 = this.config.body.modelPath) == null ? void 0 : _d2.includes("movenet"))
          bodyRes = this.config.body.enabled ? predict15(img.tensor, bodyConfig) : [];
        if (this.performance.body)
          delete this.performance.body;
      } else {
        timeStamp = now();
        if ((_e2 = this.config.body.modelPath) == null ? void 0 : _e2.includes("posenet"))
          bodyRes = this.config.body.enabled ? await predict17(img.tensor, bodyConfig) : [];
        else if ((_f = this.config.body.modelPath) == null ? void 0 : _f.includes("blazepose"))
          bodyRes = this.config.body.enabled ? await predict5(img.tensor, bodyConfig) : [];
        else if ((_g2 = this.config.body.modelPath) == null ? void 0 : _g2.includes("efficientpose"))
          bodyRes = this.config.body.enabled ? await predict7(img.tensor, bodyConfig) : [];
        else if ((_h = this.config.body.modelPath) == null ? void 0 : _h.includes("movenet"))
          bodyRes = this.config.body.enabled ? await predict15(img.tensor, bodyConfig) : [];
        this.performance.body = this.env.perfadd ? (this.performance.body || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
      }
      this.analyze("End Body:");
      this.analyze("Start Hand:");
      this.state = "detect:hand";
      const handConfig = this.config.hand.maxDetected === -1 ? mergeDeep(this.config, { hand: { maxDetected: this.config.face.enabled ? 2 * faceRes.length : 1 } }) : this.config;
      if (this.config.async) {
        if ((_j2 = (_i = this.config.hand.detector) == null ? void 0 : _i.modelPath) == null ? void 0 : _j2.includes("handdetect"))
          handRes = this.config.hand.enabled ? predict12(img.tensor, handConfig) : [];
        else if ((_l2 = (_k2 = this.config.hand.detector) == null ? void 0 : _k2.modelPath) == null ? void 0 : _l2.includes("handtrack"))
          handRes = this.config.hand.enabled ? predict13(img.tensor, handConfig) : [];
        if (this.performance.hand)
          delete this.performance.hand;
      } else {
        timeStamp = now();
        if ((_n2 = (_m2 = this.config.hand.detector) == null ? void 0 : _m2.modelPath) == null ? void 0 : _n2.includes("handdetect"))
          handRes = this.config.hand.enabled ? await predict12(img.tensor, handConfig) : [];
        else if ((_p2 = (_o2 = this.config.hand.detector) == null ? void 0 : _o2.modelPath) == null ? void 0 : _p2.includes("handtrack"))
          handRes = this.config.hand.enabled ? await predict13(img.tensor, handConfig) : [];
        this.performance.hand = this.env.perfadd ? (this.performance.hand || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
      }
      this.analyze("End Hand:");
      this.analyze("Start Object:");
      this.state = "detect:object";
      if (this.config.async) {
        if ((_q2 = this.config.object.modelPath) == null ? void 0 : _q2.includes("nanodet"))
          objectRes = this.config.object.enabled ? predict16(img.tensor, this.config) : [];
        else if ((_r2 = this.config.object.modelPath) == null ? void 0 : _r2.includes("centernet"))
          objectRes = this.config.object.enabled ? predict6(img.tensor, this.config) : [];
        if (this.performance.object)
          delete this.performance.object;
      } else {
        timeStamp = now();
        if ((_s2 = this.config.object.modelPath) == null ? void 0 : _s2.includes("nanodet"))
          objectRes = this.config.object.enabled ? await predict16(img.tensor, this.config) : [];
        else if ((_t2 = this.config.object.modelPath) == null ? void 0 : _t2.includes("centernet"))
          objectRes = this.config.object.enabled ? await predict6(img.tensor, this.config) : [];
        this.performance.object = this.env.perfadd ? (this.performance.object || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
      }
      this.analyze("End Object:");
      this.state = "detect:await";
      if (this.config.async)
        [faceRes, bodyRes, handRes, objectRes] = await Promise.all([faceRes, bodyRes, handRes, objectRes]);
      this.state = "detect:gesture";
      let gestureRes = [];
      if (this.config.gesture.enabled) {
        timeStamp = now();
        gestureRes = [...face2(faceRes), ...body2(bodyRes), ...hand2(handRes), ...iris(faceRes)];
        if (!this.config.async)
          this.performance.gesture = this.env.perfadd ? (this.performance.gesture || 0) + Math.trunc(now() - timeStamp) : Math.trunc(now() - timeStamp);
        else if (this.performance.gesture)
          delete this.performance.gesture;
      }
      this.performance.total = this.env.perfadd ? (this.performance.total || 0) + Math.trunc(now() - timeStart) : Math.trunc(now() - timeStart);
      const shape = ((_v2 = (_u2 = this.process) == null ? void 0 : _u2.tensor) == null ? void 0 : _v2.shape) || [];
      this.result = {
        face: faceRes,
        body: bodyRes,
        hand: handRes,
        gesture: gestureRes,
        object: objectRes,
        performance: this.performance,
        canvas: this.process.canvas,
        timestamp: Date.now(),
        error: null,
        get persons() {
          return join2(faceRes, bodyRes, handRes, gestureRes, shape);
        }
      };
      De(img.tensor);
      this.emit("detect");
      this.state = "idle";
      resolve(this.result);
    });
  }
};
_numTensors = new WeakMap();
_analyzeMemoryLeaks = new WeakMap();
_checkSanity = new WeakMap();
_sanity = new WeakMap();
export {
  Human,
  Human as default,
  config as defaults,
  draw_exports as draw,
  env,
  match_exports as match,
  models_exports as models
};
/**
 * @license
 * Copyright 2017 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2018 Google LLC
 *
 * Use of this source code is governed by an MIT-style
 * license that can be found in the LICENSE file or at
 * https://opensource.org/licenses/MIT.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2018 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 *
 * =============================================================================
 */
/**
 * @license
 * Copyright 2018 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2019 Google LLC
 *
 * Use of this source code is governed by an MIT-style
 * license that can be found in the LICENSE file or at
 * https://opensource.org/licenses/MIT.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2019 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 *
 * =============================================================================
 */
/**
 * @license
 * Copyright 2019 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2020 Google Inc. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2020 Google LLC
 *
 * Use of this source code is governed by an MIT-style
 * license that can be found in the LICENSE file or at
 * https://opensource.org/licenses/MIT.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2020 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use backend file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2020 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2020 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the License);
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an AS IS BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2021 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2021 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * https://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2021 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the License);
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an AS IS BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2022 Google Inc. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2022 Google LLC
 *
 * Use of this source code is governed by an MIT-style
 * license that can be found in the LICENSE file or at
 * https://opensource.org/licenses/MIT.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2022 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * @license
 * Copyright 2022 Google LLC. All Rights Reserved.
 * Licensed under the Apache License, Version 2.0 (the 'License');
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an 'AS IS' BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 * =============================================================================
 */
/**
 * Human main module
 * @default Human Library
 * @summary <https://github.com/vladmandic/human>
 * @author <https://github.com/vladmandic>
 * @copyright <https://github.com/vladmandic>
 * @license MIT
 */
/**
* @license
* Copyright 2018 Google LLC. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
* =============================================================================
*/
/** @license See the LICENSE file. */
//# sourceMappingURL=human.esm.js.map